Upload folder using huggingface_hub

exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/stats.1.log

@@ -0,0 +1,874 @@
+# python3 -m espnet2.bin.gan_tts_train --collect_stats true --write_collected_feats false --use_preprocessor true --token_type phn --token_list dump/token_list/phn_jaconv_pyopenjtalk/tokens.txt --non_linguistic_symbols none --cleaner jaconv --g2p pyopenjtalk --normalize none --pitch_normalize none --energy_normalize none --train_data_path_and_name_and_type dump/raw/jvs010_tr_no_dev/text,text,text --train_data_path_and_name_and_type dump/raw/jvs010_tr_no_dev/wav.scp,speech,sound --valid_data_path_and_name_and_type dump/raw/jvs010_dev/text,text,text --valid_data_path_and_name_and_type dump/raw/jvs010_dev/wav.scp,speech,sound --train_shape_file exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/train.1.scp --valid_shape_file exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/valid.1.scp --output_dir exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/stats.1 --config conf/tuning/train_jets.yaml --feats_extract fbank --feats_extract_conf n_fft=2048 --feats_extract_conf hop_length=300 --feats_extract_conf win_length=1200 --feats_extract_conf fs=24000 --feats_extract_conf fmin=80 --feats_extract_conf fmax=7600 --feats_extract_conf n_mels=80 --pitch_extract_conf fs=24000 --pitch_extract_conf n_fft=2048 --pitch_extract_conf hop_length=300 --pitch_extract_conf f0max=400 --pitch_extract_conf f0min=80 --energy_extract_conf fs=24000 --energy_extract_conf n_fft=2048 --energy_extract_conf hop_length=300 --energy_extract_conf win_length=1200 
+# Started at Tue Mar  4 21:23:26 JST 2025
+#
+/usr/lib/python3/dist-packages/requests/__init__.py:89: RequestsDependencyWarning: urllib3 (2.2.3) or chardet (3.0.4) doesn't match a supported version!
+  warnings.warn("urllib3 ({}) or chardet ({}) doesn't match a supported "
+/usr/bin/python3 /work/espnet/espnet2/bin/gan_tts_train.py --collect_stats true --write_collected_feats false --use_preprocessor true --token_type phn --token_list dump/token_list/phn_jaconv_pyopenjtalk/tokens.txt --non_linguistic_symbols none --cleaner jaconv --g2p pyopenjtalk --normalize none --pitch_normalize none --energy_normalize none --train_data_path_and_name_and_type dump/raw/jvs010_tr_no_dev/text,text,text --train_data_path_and_name_and_type dump/raw/jvs010_tr_no_dev/wav.scp,speech,sound --valid_data_path_and_name_and_type dump/raw/jvs010_dev/text,text,text --valid_data_path_and_name_and_type dump/raw/jvs010_dev/wav.scp,speech,sound --train_shape_file exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/train.1.scp --valid_shape_file exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/valid.1.scp --output_dir exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/stats.1 --config conf/tuning/train_jets.yaml --feats_extract fbank --feats_extract_conf n_fft=2048 --feats_extract_conf hop_length=300 --feats_extract_conf win_length=1200 --feats_extract_conf fs=24000 --feats_extract_conf fmin=80 --feats_extract_conf fmax=7600 --feats_extract_conf n_mels=80 --pitch_extract_conf fs=24000 --pitch_extract_conf n_fft=2048 --pitch_extract_conf hop_length=300 --pitch_extract_conf f0max=400 --pitch_extract_conf f0min=80 --energy_extract_conf fs=24000 --energy_extract_conf n_fft=2048 --energy_extract_conf hop_length=300 --energy_extract_conf win_length=1200
+[92b100c97f43] 2025-03-04 21:23:29,215 (gan_tts:304) INFO: Vocabulary size: 41
+[92b100c97f43] 2025-03-04 21:23:29,440 (encoder:172) INFO: encoder self-attention layer type = self-attention
+[92b100c97f43] 2025-03-04 21:23:29,563 (encoder:172) INFO: encoder self-attention layer type = self-attention
+[92b100c97f43] 2025-03-04 21:23:31,676 (abs_task:1157) INFO: pytorch.version=1.10.1+cu113, cuda.available=True, cudnn.version=8200, cudnn.benchmark=False, cudnn.deterministic=False
+[92b100c97f43] 2025-03-04 21:23:31,686 (abs_task:1158) INFO: Model structure:
+ESPnetGANTTSModel(
+  (feats_extract): LogMelFbank(
+    (stft): Stft(n_fft=2048, win_length=1200, hop_length=300, center=True, normalized=False, onesided=True)
+    (logmel): LogMel(sr=24000, n_fft=2048, n_mels=80, fmin=80, fmax=7600, htk=False)
+  )
+  (pitch_extract): Dio()
+  (energy_extract): Energy(
+    (stft): Stft(n_fft=2048, win_length=1200, hop_length=300, center=True, normalized=False, onesided=True)
+  )
+  (tts): JETS(
+    (generator): JETSGenerator(
+      (encoder): Encoder(
+        (embed): Sequential(
+          (0): Embedding(41, 256, padding_idx=0)
+          (1): ScaledPositionalEncoding(
+            (dropout): Dropout(p=0.2, inplace=False)
+          )
+        )
+        (encoders): MultiSequential(
+          (0): EncoderLayer(
+            (self_attn): MultiHeadedAttention(
+              (linear_q): Linear(in_features=256, out_features=256, bias=True)
+              (linear_k): Linear(in_features=256, out_features=256, bias=True)
+              (linear_v): Linear(in_features=256, out_features=256, bias=True)
+              (linear_out): Linear(in_features=256, out_features=256, bias=True)
+              (dropout): Dropout(p=0.2, inplace=False)
+            )
+            (feed_forward): MultiLayeredConv1d(
+              (w_1): Conv1d(256, 1024, kernel_size=(3,), stride=(1,), padding=(1,))
+              (w_2): Conv1d(1024, 256, kernel_size=(3,), stride=(1,), padding=(1,))
+              (dropout): Dropout(p=0.2, inplace=False)
+            )
+            (norm1): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (norm2): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (dropout): Dropout(p=0.2, inplace=False)
+          )
+          (1): EncoderLayer(
+            (self_attn): MultiHeadedAttention(
+              (linear_q): Linear(in_features=256, out_features=256, bias=True)
+              (linear_k): Linear(in_features=256, out_features=256, bias=True)
+              (linear_v): Linear(in_features=256, out_features=256, bias=True)
+              (linear_out): Linear(in_features=256, out_features=256, bias=True)
+              (dropout): Dropout(p=0.2, inplace=False)
+            )
+            (feed_forward): MultiLayeredConv1d(
+              (w_1): Conv1d(256, 1024, kernel_size=(3,), stride=(1,), padding=(1,))
+              (w_2): Conv1d(1024, 256, kernel_size=(3,), stride=(1,), padding=(1,))
+              (dropout): Dropout(p=0.2, inplace=False)
+            )
+            (norm1): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (norm2): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (dropout): Dropout(p=0.2, inplace=False)
+          )
+          (2): EncoderLayer(
+            (self_attn): MultiHeadedAttention(
+              (linear_q): Linear(in_features=256, out_features=256, bias=True)
+              (linear_k): Linear(in_features=256, out_features=256, bias=True)
+              (linear_v): Linear(in_features=256, out_features=256, bias=True)
+              (linear_out): Linear(in_features=256, out_features=256, bias=True)
+              (dropout): Dropout(p=0.2, inplace=False)
+            )
+            (feed_forward): MultiLayeredConv1d(
+              (w_1): Conv1d(256, 1024, kernel_size=(3,), stride=(1,), padding=(1,))
+              (w_2): Conv1d(1024, 256, kernel_size=(3,), stride=(1,), padding=(1,))
+              (dropout): Dropout(p=0.2, inplace=False)
+            )
+            (norm1): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (norm2): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (dropout): Dropout(p=0.2, inplace=False)
+          )
+          (3): EncoderLayer(
+            (self_attn): MultiHeadedAttention(
+              (linear_q): Linear(in_features=256, out_features=256, bias=True)
+              (linear_k): Linear(in_features=256, out_features=256, bias=True)
+              (linear_v): Linear(in_features=256, out_features=256, bias=True)
+              (linear_out): Linear(in_features=256, out_features=256, bias=True)
+              (dropout): Dropout(p=0.2, inplace=False)
+            )
+            (feed_forward): MultiLayeredConv1d(
+              (w_1): Conv1d(256, 1024, kernel_size=(3,), stride=(1,), padding=(1,))
+              (w_2): Conv1d(1024, 256, kernel_size=(3,), stride=(1,), padding=(1,))
+              (dropout): Dropout(p=0.2, inplace=False)
+            )
+            (norm1): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (norm2): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (dropout): Dropout(p=0.2, inplace=False)
+          )
+        )
+        (after_norm): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+      )
+      (duration_predictor): DurationPredictor(
+        (conv): ModuleList(
+          (0): Sequential(
+            (0): Conv1d(256, 256, kernel_size=(3,), stride=(1,), padding=(1,))
+            (1): ReLU()
+            (2): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (3): Dropout(p=0.1, inplace=False)
+          )
+          (1): Sequential(
+            (0): Conv1d(256, 256, kernel_size=(3,), stride=(1,), padding=(1,))
+            (1): ReLU()
+            (2): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (3): Dropout(p=0.1, inplace=False)
+          )
+        )
+        (linear): Linear(in_features=256, out_features=1, bias=True)
+      )
+      (pitch_predictor): VariancePredictor(
+        (conv): ModuleList(
+          (0): Sequential(
+            (0): Conv1d(256, 256, kernel_size=(5,), stride=(1,), padding=(2,))
+            (1): ReLU()
+            (2): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (3): Dropout(p=0.5, inplace=False)
+          )
+          (1): Sequential(
+            (0): Conv1d(256, 256, kernel_size=(5,), stride=(1,), padding=(2,))
+            (1): ReLU()
+            (2): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (3): Dropout(p=0.5, inplace=False)
+          )
+          (2): Sequential(
+            (0): Conv1d(256, 256, kernel_size=(5,), stride=(1,), padding=(2,))
+            (1): ReLU()
+            (2): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (3): Dropout(p=0.5, inplace=False)
+          )
+          (3): Sequential(
+            (0): Conv1d(256, 256, kernel_size=(5,), stride=(1,), padding=(2,))
+            (1): ReLU()
+            (2): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (3): Dropout(p=0.5, inplace=False)
+          )
+          (4): Sequential(
+            (0): Conv1d(256, 256, kernel_size=(5,), stride=(1,), padding=(2,))
+            (1): ReLU()
+            (2): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (3): Dropout(p=0.5, inplace=False)
+          )
+        )
+        (linear): Linear(in_features=256, out_features=1, bias=True)
+      )
+      (pitch_embed): Sequential(
+        (0): Conv1d(1, 256, kernel_size=(1,), stride=(1,))
+        (1): Dropout(p=0.0, inplace=False)
+      )
+      (energy_predictor): VariancePredictor(
+        (conv): ModuleList(
+          (0): Sequential(
+            (0): Conv1d(256, 256, kernel_size=(3,), stride=(1,), padding=(1,))
+            (1): ReLU()
+            (2): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (3): Dropout(p=0.5, inplace=False)
+          )
+          (1): Sequential(
+            (0): Conv1d(256, 256, kernel_size=(3,), stride=(1,), padding=(1,))
+            (1): ReLU()
+            (2): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (3): Dropout(p=0.5, inplace=False)
+          )
+        )
+        (linear): Linear(in_features=256, out_features=1, bias=True)
+      )
+      (energy_embed): Sequential(
+        (0): Conv1d(1, 256, kernel_size=(1,), stride=(1,))
+        (1): Dropout(p=0.0, inplace=False)
+      )
+      (alignment_module): AlignmentModule(
+        (t_conv1): Conv1d(256, 256, kernel_size=(3,), stride=(1,), padding=(1,))
+        (t_conv2): Conv1d(256, 256, kernel_size=(1,), stride=(1,))
+        (f_conv1): Conv1d(80, 256, kernel_size=(3,), stride=(1,), padding=(1,))
+        (f_conv2): Conv1d(256, 256, kernel_size=(3,), stride=(1,), padding=(1,))
+        (f_conv3): Conv1d(256, 256, kernel_size=(1,), stride=(1,))
+      )
+      (length_regulator): GaussianUpsampling()
+      (decoder): Encoder(
+        (embed): Sequential(
+          (0): ScaledPositionalEncoding(
+            (dropout): Dropout(p=0.2, inplace=False)
+          )
+        )
+        (encoders): MultiSequential(
+          (0): EncoderLayer(
+            (self_attn): MultiHeadedAttention(
+              (linear_q): Linear(in_features=256, out_features=256, bias=True)
+              (linear_k): Linear(in_features=256, out_features=256, bias=True)
+              (linear_v): Linear(in_features=256, out_features=256, bias=True)
+              (linear_out): Linear(in_features=256, out_features=256, bias=True)
+              (dropout): Dropout(p=0.2, inplace=False)
+            )
+            (feed_forward): MultiLayeredConv1d(
+              (w_1): Conv1d(256, 1024, kernel_size=(3,), stride=(1,), padding=(1,))
+              (w_2): Conv1d(1024, 256, kernel_size=(3,), stride=(1,), padding=(1,))
+              (dropout): Dropout(p=0.2, inplace=False)
+            )
+            (norm1): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (norm2): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (dropout): Dropout(p=0.2, inplace=False)
+          )
+          (1): EncoderLayer(
+            (self_attn): MultiHeadedAttention(
+              (linear_q): Linear(in_features=256, out_features=256, bias=True)
+              (linear_k): Linear(in_features=256, out_features=256, bias=True)
+              (linear_v): Linear(in_features=256, out_features=256, bias=True)
+              (linear_out): Linear(in_features=256, out_features=256, bias=True)
+              (dropout): Dropout(p=0.2, inplace=False)
+            )
+            (feed_forward): MultiLayeredConv1d(
+              (w_1): Conv1d(256, 1024, kernel_size=(3,), stride=(1,), padding=(1,))
+              (w_2): Conv1d(1024, 256, kernel_size=(3,), stride=(1,), padding=(1,))
+              (dropout): Dropout(p=0.2, inplace=False)
+            )
+            (norm1): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (norm2): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (dropout): Dropout(p=0.2, inplace=False)
+          )
+          (2): EncoderLayer(
+            (self_attn): MultiHeadedAttention(
+              (linear_q): Linear(in_features=256, out_features=256, bias=True)
+              (linear_k): Linear(in_features=256, out_features=256, bias=True)
+              (linear_v): Linear(in_features=256, out_features=256, bias=True)
+              (linear_out): Linear(in_features=256, out_features=256, bias=True)
+              (dropout): Dropout(p=0.2, inplace=False)
+            )
+            (feed_forward): MultiLayeredConv1d(
+              (w_1): Conv1d(256, 1024, kernel_size=(3,), stride=(1,), padding=(1,))
+              (w_2): Conv1d(1024, 256, kernel_size=(3,), stride=(1,), padding=(1,))
+              (dropout): Dropout(p=0.2, inplace=False)
+            )
+            (norm1): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (norm2): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (dropout): Dropout(p=0.2, inplace=False)
+          )
+          (3): EncoderLayer(
+            (self_attn): MultiHeadedAttention(
+              (linear_q): Linear(in_features=256, out_features=256, bias=True)
+              (linear_k): Linear(in_features=256, out_features=256, bias=True)
+              (linear_v): Linear(in_features=256, out_features=256, bias=True)
+              (linear_out): Linear(in_features=256, out_features=256, bias=True)
+              (dropout): Dropout(p=0.2, inplace=False)
+            )
+            (feed_forward): MultiLayeredConv1d(
+              (w_1): Conv1d(256, 1024, kernel_size=(3,), stride=(1,), padding=(1,))
+              (w_2): Conv1d(1024, 256, kernel_size=(3,), stride=(1,), padding=(1,))
+              (dropout): Dropout(p=0.2, inplace=False)
+            )
+            (norm1): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (norm2): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (dropout): Dropout(p=0.2, inplace=False)
+          )
+        )
+        (after_norm): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+      )
+      (generator): HiFiGANGenerator(
+        (input_conv): Conv1d(256, 512, kernel_size=(7,), stride=(1,), padding=(3,))
+        (upsamples): ModuleList(
+          (0): Sequential(
+            (0): LeakyReLU(negative_slope=0.1)
+            (1): ConvTranspose1d(512, 256, kernel_size=(16,), stride=(8,), padding=(4,))
+          )
+          (1): Sequential(
+            (0): LeakyReLU(negative_slope=0.1)
+            (1): ConvTranspose1d(256, 128, kernel_size=(16,), stride=(8,), padding=(4,))
+          )
+          (2): Sequential(
+            (0): LeakyReLU(negative_slope=0.1)
+            (1): ConvTranspose1d(128, 64, kernel_size=(4,), stride=(2,), padding=(1,))
+          )
+          (3): Sequential(
+            (0): LeakyReLU(negative_slope=0.1)
+            (1): ConvTranspose1d(64, 32, kernel_size=(4,), stride=(2,), padding=(1,))
+          )
+        )
+        (blocks): ModuleList(
+          (0): ResidualBlock(
+            (convs1): ModuleList(
+              (0): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(256, 256, kernel_size=(3,), stride=(1,), padding=(1,))
+              )
+              (1): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(256, 256, kernel_size=(3,), stride=(1,), padding=(3,), dilation=(3,))
+              )
+              (2): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(256, 256, kernel_size=(3,), stride=(1,), padding=(5,), dilation=(5,))
+              )
+            )
+            (convs2): ModuleList(
+              (0): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(256, 256, kernel_size=(3,), stride=(1,), padding=(1,))
+              )
+              (1): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(256, 256, kernel_size=(3,), stride=(1,), padding=(1,))
+              )
+              (2): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(256, 256, kernel_size=(3,), stride=(1,), padding=(1,))
+              )
+            )
+          )
+          (1): ResidualBlock(
+            (convs1): ModuleList(
+              (0): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(256, 256, kernel_size=(7,), stride=(1,), padding=(3,))
+              )
+              (1): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(256, 256, kernel_size=(7,), stride=(1,), padding=(9,), dilation=(3,))
+              )
+              (2): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(256, 256, kernel_size=(7,), stride=(1,), padding=(15,), dilation=(5,))
+              )
+            )
+            (convs2): ModuleList(
+              (0): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(256, 256, kernel_size=(7,), stride=(1,), padding=(3,))
+              )
+              (1): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(256, 256, kernel_size=(7,), stride=(1,), padding=(3,))
+              )
+              (2): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(256, 256, kernel_size=(7,), stride=(1,), padding=(3,))
+              )
+            )
+          )
+          (2): ResidualBlock(
+            (convs1): ModuleList(
+              (0): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(256, 256, kernel_size=(11,), stride=(1,), padding=(5,))
+              )
+              (1): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(256, 256, kernel_size=(11,), stride=(1,), padding=(15,), dilation=(3,))
+              )
+              (2): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(256, 256, kernel_size=(11,), stride=(1,), padding=(25,), dilation=(5,))
+              )
+            )
+            (convs2): ModuleList(
+              (0): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(256, 256, kernel_size=(11,), stride=(1,), padding=(5,))
+              )
+              (1): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(256, 256, kernel_size=(11,), stride=(1,), padding=(5,))
+              )
+              (2): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(256, 256, kernel_size=(11,), stride=(1,), padding=(5,))
+              )
+            )
+          )
+          (3): ResidualBlock(
+            (convs1): ModuleList(
+              (0): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(128, 128, kernel_size=(3,), stride=(1,), padding=(1,))
+              )
+              (1): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(128, 128, kernel_size=(3,), stride=(1,), padding=(3,), dilation=(3,))
+              )
+              (2): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(128, 128, kernel_size=(3,), stride=(1,), padding=(5,), dilation=(5,))
+              )
+            )
+            (convs2): ModuleList(
+              (0): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(128, 128, kernel_size=(3,), stride=(1,), padding=(1,))
+              )
+              (1): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(128, 128, kernel_size=(3,), stride=(1,), padding=(1,))
+              )
+              (2): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(128, 128, kernel_size=(3,), stride=(1,), padding=(1,))
+              )
+            )
+          )
+          (4): ResidualBlock(
+            (convs1): ModuleList(
+              (0): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(128, 128, kernel_size=(7,), stride=(1,), padding=(3,))
+              )
+              (1): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(128, 128, kernel_size=(7,), stride=(1,), padding=(9,), dilation=(3,))
+              )
+              (2): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(128, 128, kernel_size=(7,), stride=(1,), padding=(15,), dilation=(5,))
+              )
+            )
+            (convs2): ModuleList(
+              (0): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(128, 128, kernel_size=(7,), stride=(1,), padding=(3,))
+              )
+              (1): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(128, 128, kernel_size=(7,), stride=(1,), padding=(3,))
+              )
+              (2): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(128, 128, kernel_size=(7,), stride=(1,), padding=(3,))
+              )
+            )
+          )
+          (5): ResidualBlock(
+            (convs1): ModuleList(
+              (0): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(128, 128, kernel_size=(11,), stride=(1,), padding=(5,))
+              )
+              (1): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(128, 128, kernel_size=(11,), stride=(1,), padding=(15,), dilation=(3,))
+              )
+              (2): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(128, 128, kernel_size=(11,), stride=(1,), padding=(25,), dilation=(5,))
+              )
+            )
+            (convs2): ModuleList(
+              (0): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(128, 128, kernel_size=(11,), stride=(1,), padding=(5,))
+              )
+              (1): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(128, 128, kernel_size=(11,), stride=(1,), padding=(5,))
+              )
+              (2): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(128, 128, kernel_size=(11,), stride=(1,), padding=(5,))
+              )
+            )
+          )
+          (6): ResidualBlock(
+            (convs1): ModuleList(
+              (0): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(64, 64, kernel_size=(3,), stride=(1,), padding=(1,))
+              )
+              (1): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(64, 64, kernel_size=(3,), stride=(1,), padding=(3,), dilation=(3,))
+              )
+              (2): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(64, 64, kernel_size=(3,), stride=(1,), padding=(5,), dilation=(5,))
+              )
+            )
+            (convs2): ModuleList(
+              (0): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(64, 64, kernel_size=(3,), stride=(1,), padding=(1,))
+              )
+              (1): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(64, 64, kernel_size=(3,), stride=(1,), padding=(1,))
+              )
+              (2): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(64, 64, kernel_size=(3,), stride=(1,), padding=(1,))
+              )
+            )
+          )
+          (7): ResidualBlock(
+            (convs1): ModuleList(
+              (0): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(64, 64, kernel_size=(7,), stride=(1,), padding=(3,))
+              )
+              (1): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(64, 64, kernel_size=(7,), stride=(1,), padding=(9,), dilation=(3,))
+              )
+              (2): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(64, 64, kernel_size=(7,), stride=(1,), padding=(15,), dilation=(5,))
+              )
+            )
+            (convs2): ModuleList(
+              (0): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(64, 64, kernel_size=(7,), stride=(1,), padding=(3,))
+              )
+              (1): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(64, 64, kernel_size=(7,), stride=(1,), padding=(3,))
+              )
+              (2): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(64, 64, kernel_size=(7,), stride=(1,), padding=(3,))
+              )
+            )
+          )
+          (8): ResidualBlock(
+            (convs1): ModuleList(
+              (0): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(64, 64, kernel_size=(11,), stride=(1,), padding=(5,))
+              )
+              (1): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(64, 64, kernel_size=(11,), stride=(1,), padding=(15,), dilation=(3,))
+              )
+              (2): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(64, 64, kernel_size=(11,), stride=(1,), padding=(25,), dilation=(5,))
+              )
+            )
+            (convs2): ModuleList(
+              (0): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(64, 64, kernel_size=(11,), stride=(1,), padding=(5,))
+              )
+              (1): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(64, 64, kernel_size=(11,), stride=(1,), padding=(5,))
+              )
+              (2): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(64, 64, kernel_size=(11,), stride=(1,), padding=(5,))
+              )
+            )
+          )
+          (9): ResidualBlock(
+            (convs1): ModuleList(
+              (0): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(32, 32, kernel_size=(3,), stride=(1,), padding=(1,))
+              )
+              (1): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(32, 32, kernel_size=(3,), stride=(1,), padding=(3,), dilation=(3,))
+              )
+              (2): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(32, 32, kernel_size=(3,), stride=(1,), padding=(5,), dilation=(5,))
+              )
+            )
+            (convs2): ModuleList(
+              (0): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(32, 32, kernel_size=(3,), stride=(1,), padding=(1,))
+              )
+              (1): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(32, 32, kernel_size=(3,), stride=(1,), padding=(1,))
+              )
+              (2): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(32, 32, kernel_size=(3,), stride=(1,), padding=(1,))
+              )
+            )
+          )
+          (10): ResidualBlock(
+            (convs1): ModuleList(
+              (0): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(32, 32, kernel_size=(7,), stride=(1,), padding=(3,))
+              )
+              (1): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(32, 32, kernel_size=(7,), stride=(1,), padding=(9,), dilation=(3,))
+              )
+              (2): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(32, 32, kernel_size=(7,), stride=(1,), padding=(15,), dilation=(5,))
+              )
+            )
+            (convs2): ModuleList(
+              (0): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(32, 32, kernel_size=(7,), stride=(1,), padding=(3,))
+              )
+              (1): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(32, 32, kernel_size=(7,), stride=(1,), padding=(3,))
+              )
+              (2): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(32, 32, kernel_size=(7,), stride=(1,), padding=(3,))
+              )
+            )
+          )
+          (11): ResidualBlock(
+            (convs1): ModuleList(
+              (0): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(32, 32, kernel_size=(11,), stride=(1,), padding=(5,))
+              )
+              (1): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(32, 32, kernel_size=(11,), stride=(1,), padding=(15,), dilation=(3,))
+              )
+              (2): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(32, 32, kernel_size=(11,), stride=(1,), padding=(25,), dilation=(5,))
+              )
+            )
+            (convs2): ModuleList(
+              (0): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(32, 32, kernel_size=(11,), stride=(1,), padding=(5,))
+              )
+              (1): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(32, 32, kernel_size=(11,), stride=(1,), padding=(5,))
+              )
+              (2): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(32, 32, kernel_size=(11,), stride=(1,), padding=(5,))
+              )
+            )
+          )
+        )
+        (output_conv): Sequential(
+          (0): LeakyReLU(negative_slope=0.01)
+          (1): Conv1d(32, 1, kernel_size=(7,), stride=(1,), padding=(3,))
+          (2): Tanh()
+        )
+      )
+    )
+    (discriminator): HiFiGANMultiScaleMultiPeriodDiscriminator(
+      (msd): HiFiGANMultiScaleDiscriminator(
+        (discriminators): ModuleList(
+          (0): HiFiGANScaleDiscriminator(
+            (layers): ModuleList(
+              (0): Sequential(
+                (0): Conv1d(1, 128, kernel_size=(15,), stride=(1,), padding=(7,))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (1): Sequential(
+                (0): Conv1d(128, 128, kernel_size=(41,), stride=(2,), padding=(20,), groups=4)
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (2): Sequential(
+                (0): Conv1d(128, 256, kernel_size=(41,), stride=(2,), padding=(20,), groups=16)
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (3): Sequential(
+                (0): Conv1d(256, 512, kernel_size=(41,), stride=(4,), padding=(20,), groups=16)
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (4): Sequential(
+                (0): Conv1d(512, 1024, kernel_size=(41,), stride=(4,), padding=(20,), groups=16)
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (5): Sequential(
+                (0): Conv1d(1024, 1024, kernel_size=(41,), stride=(1,), padding=(20,), groups=16)
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (6): Sequential(
+                (0): Conv1d(1024, 1024, kernel_size=(5,), stride=(1,), padding=(2,))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (7): Conv1d(1024, 1, kernel_size=(3,), stride=(1,), padding=(1,))
+            )
+          )
+        )
+      )
+      (mpd): HiFiGANMultiPeriodDiscriminator(
+        (discriminators): ModuleList(
+          (0): HiFiGANPeriodDiscriminator(
+            (convs): ModuleList(
+              (0): Sequential(
+                (0): Conv2d(1, 32, kernel_size=(5, 1), stride=(3, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (1): Sequential(
+                (0): Conv2d(32, 128, kernel_size=(5, 1), stride=(3, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (2): Sequential(
+                (0): Conv2d(128, 512, kernel_size=(5, 1), stride=(3, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (3): Sequential(
+                (0): Conv2d(512, 1024, kernel_size=(5, 1), stride=(3, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (4): Sequential(
+                (0): Conv2d(1024, 1024, kernel_size=(5, 1), stride=(1, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+            )
+            (output_conv): Conv2d(1024, 1, kernel_size=(2, 1), stride=(1, 1), padding=(1, 0))
+          )
+          (1): HiFiGANPeriodDiscriminator(
+            (convs): ModuleList(
+              (0): Sequential(
+                (0): Conv2d(1, 32, kernel_size=(5, 1), stride=(3, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (1): Sequential(
+                (0): Conv2d(32, 128, kernel_size=(5, 1), stride=(3, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (2): Sequential(
+                (0): Conv2d(128, 512, kernel_size=(5, 1), stride=(3, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (3): Sequential(
+                (0): Conv2d(512, 1024, kernel_size=(5, 1), stride=(3, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (4): Sequential(
+                (0): Conv2d(1024, 1024, kernel_size=(5, 1), stride=(1, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+            )
+            (output_conv): Conv2d(1024, 1, kernel_size=(2, 1), stride=(1, 1), padding=(1, 0))
+          )
+          (2): HiFiGANPeriodDiscriminator(
+            (convs): ModuleList(
+              (0): Sequential(
+                (0): Conv2d(1, 32, kernel_size=(5, 1), stride=(3, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (1): Sequential(
+                (0): Conv2d(32, 128, kernel_size=(5, 1), stride=(3, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (2): Sequential(
+                (0): Conv2d(128, 512, kernel_size=(5, 1), stride=(3, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (3): Sequential(
+                (0): Conv2d(512, 1024, kernel_size=(5, 1), stride=(3, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (4): Sequential(
+                (0): Conv2d(1024, 1024, kernel_size=(5, 1), stride=(1, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+            )
+            (output_conv): Conv2d(1024, 1, kernel_size=(2, 1), stride=(1, 1), padding=(1, 0))
+          )
+          (3): HiFiGANPeriodDiscriminator(
+            (convs): ModuleList(
+              (0): Sequential(
+                (0): Conv2d(1, 32, kernel_size=(5, 1), stride=(3, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (1): Sequential(
+                (0): Conv2d(32, 128, kernel_size=(5, 1), stride=(3, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (2): Sequential(
+                (0): Conv2d(128, 512, kernel_size=(5, 1), stride=(3, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (3): Sequential(
+                (0): Conv2d(512, 1024, kernel_size=(5, 1), stride=(3, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (4): Sequential(
+                (0): Conv2d(1024, 1024, kernel_size=(5, 1), stride=(1, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+            )
+            (output_conv): Conv2d(1024, 1, kernel_size=(2, 1), stride=(1, 1), padding=(1, 0))
+          )
+          (4): HiFiGANPeriodDiscriminator(
+            (convs): ModuleList(
+              (0): Sequential(
+                (0): Conv2d(1, 32, kernel_size=(5, 1), stride=(3, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (1): Sequential(
+                (0): Conv2d(32, 128, kernel_size=(5, 1), stride=(3, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (2): Sequential(
+                (0): Conv2d(128, 512, kernel_size=(5, 1), stride=(3, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (3): Sequential(
+                (0): Conv2d(512, 1024, kernel_size=(5, 1), stride=(3, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (4): Sequential(
+                (0): Conv2d(1024, 1024, kernel_size=(5, 1), stride=(1, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+            )
+            (output_conv): Conv2d(1024, 1, kernel_size=(2, 1), stride=(1, 1), padding=(1, 0))
+          )
+        )
+      )
+    )
+    (generator_adv_loss): GeneratorAdversarialLoss()
+    (discriminator_adv_loss): DiscriminatorAdversarialLoss()
+    (feat_match_loss): FeatureMatchLoss()
+    (mel_loss): MelSpectrogramLoss(
+      (wav_to_mel): LogMelFbank(
+        (stft): Stft(n_fft=1024, win_length=1024, hop_length=256, center=True, normalized=False, onesided=True)
+        (logmel): LogMel(sr=24000, n_fft=1024, n_mels=80, fmin=0, fmax=12000.0, htk=False)
+      )
+    )
+    (var_loss): VarianceLoss(
+      (mse_criterion): MSELoss()
+      (duration_criterion): DurationPredictorLoss(
+        (criterion): MSELoss()
+      )
+    )
+    (forwardsum_loss): ForwardSumLoss()
+  )
+)
+
+Model summary:
+    Class Name: ESPnetGANTTSModel
+    Total Number of model parameters: 83.28 M
+    Number of trainable parameters: 83.28 M (100.0%)
+    Size: 333.11 MB
+    Type: torch.float32
+[92b100c97f43] 2025-03-04 21:23:31,686 (abs_task:1161) INFO: Optimizer:
+AdamW (
+Parameter Group 0
+    amsgrad: False
+    betas: [0.8, 0.99]
+    eps: 1e-09
+    initial_lr: 0.0002
+    lr: 0.0002
+    weight_decay: 0.0
+)
+[92b100c97f43] 2025-03-04 21:23:31,686 (abs_task:1162) INFO: Scheduler: <torch.optim.lr_scheduler.ExponentialLR object at 0x7fea878fb160>
+[92b100c97f43] 2025-03-04 21:23:31,686 (abs_task:1161) INFO: Optimizer2:
+AdamW (
+Parameter Group 0
+    amsgrad: False
+    betas: [0.8, 0.99]
+    eps: 1e-09
+    initial_lr: 0.0002
+    lr: 0.0002
+    weight_decay: 0.0
+)
+[92b100c97f43] 2025-03-04 21:23:31,686 (abs_task:1162) INFO: Scheduler2: <torch.optim.lr_scheduler.ExponentialLR object at 0x7fea9681f280>
+[92b100c97f43] 2025-03-04 21:23:31,686 (abs_task:1171) INFO: Saving the configuration in exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/stats.1/config.yaml
+[92b100c97f43] 2025-03-04 21:23:31,705 (abs_task:1182) INFO: Namespace(accum_grad=1, allow_variable_data_keys=False, batch_bins=6000000, batch_size=20, batch_type='numel', best_model_criterion=[['valid', 'text2mel_loss', 'min'], ['train', 'text2mel_loss', 'min'], ['train', 'total_count', 'max']], bpemodel=None, chunk_length=500, chunk_shift_ratio=0.5, cleaner='jaconv', collect_stats=True, config='conf/tuning/train_jets.yaml', cudnn_benchmark=False, cudnn_deterministic=False, cudnn_enabled=True, detect_anomaly=False, dist_backend='nccl', dist_init_method='env://', dist_launcher=None, dist_master_addr=None, dist_master_port=None, dist_rank=None, dist_world_size=None, distributed=False, dry_run=False, early_stopping_criterion=('valid', 'loss', 'min'), energy_extract='energy', energy_extract_conf={'reduction_factor': 1, 'use_token_averaged_energy': False, 'fs': 24000, 'n_fft': 2048, 'hop_length': 300, 'win_length': 1200}, energy_normalize=None, energy_normalize_conf={}, feats_extract='fbank', feats_extract_conf={'n_fft': 2048, 'hop_length': 300, 'win_length': 1200, 'fs': 24000, 'fmin': 80, 'fmax': 7600, 'n_mels': 80}, fold_length=[], freeze_param=[], g2p='pyopenjtalk', generator_first=True, grad_clip=-1, grad_clip_type=2.0, grad_noise=False, ignore_init_mismatch=False, init_param=[], iterator_type='sequence', keep_nbest_models=-1, local_rank=None, log_interval=50, log_level='INFO', max_cache_fd=32, max_cache_size=0.0, max_epoch=130, model_conf={}, multiple_iterator=False, multiprocessing_distributed=False, nbest_averaging_interval=0, ngpu=0, no_forward_run=False, non_linguistic_symbols=None, normalize=None, normalize_conf={}, num_att_plot=3, num_cache_chunks=1024, num_iters_per_epoch=1000, num_workers=16, odim=None, optim='adamw', optim2='adamw', optim2_conf={'lr': 0.0002, 'betas': [0.8, 0.99], 'eps': 1e-09, 'weight_decay': 0.0}, optim_conf={'lr': 0.0002, 'betas': [0.8, 0.99], 'eps': 1e-09, 'weight_decay': 0.0}, output_dir='exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/stats.1', patience=None, pitch_extract='dio', pitch_extract_conf={'reduction_factor': 1, 'use_token_averaged_f0': False, 'fs': 24000, 'n_fft': 2048, 'hop_length': 300, 'f0max': 400, 'f0min': 80}, pitch_normalize=None, pitch_normalize_conf={}, pretrain_path=None, print_config=False, required=['output_dir', 'token_list'], resume=False, scheduler='exponentiallr', scheduler2='exponentiallr', scheduler2_conf={'gamma': 0.999875}, scheduler_conf={'gamma': 0.999875}, seed=777, sharded_ddp=False, sort_batch='descending', sort_in_batch='descending', token_list=['<blank>', '<unk>', 'o', 'a', 'u', 'i', 'e', 'k', 'r', 't', 'n', 'pau', 'N', 's', 'sh', 'd', 'm', 'g', 'w', 'b', 'cl', 'I', 'j', 'ch', 'y', 'U', 'h', 'p', 'ts', 'f', 'z', 'ky', 'ny', 'gy', 'ry', 'hy', 'my', 'by', 'py', 'v', '<sos/eos>'], token_type='phn', train_data_path_and_name_and_type=[('dump/raw/jvs010_tr_no_dev/text', 'text', 'text'), ('dump/raw/jvs010_tr_no_dev/wav.scp', 'speech', 'sound')], train_dtype='float32', train_shape_file=['exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/train.1.scp'], tts='jets', tts_conf={'generator_type': 'jets_generator', 'generator_params': {'adim': 256, 'aheads': 2, 'elayers': 4, 'eunits': 1024, 'dlayers': 4, 'dunits': 1024, 'positionwise_layer_type': 'conv1d', 'positionwise_conv_kernel_size': 3, 'duration_predictor_layers': 2, 'duration_predictor_chans': 256, 'duration_predictor_kernel_size': 3, 'use_masking': True, 'encoder_normalize_before': True, 'decoder_normalize_before': True, 'encoder_type': 'transformer', 'decoder_type': 'transformer', 'conformer_rel_pos_type': 'latest', 'conformer_pos_enc_layer_type': 'rel_pos', 'conformer_self_attn_layer_type': 'rel_selfattn', 'conformer_activation_type': 'swish', 'use_macaron_style_in_conformer': True, 'use_cnn_in_conformer': True, 'conformer_enc_kernel_size': 7, 'conformer_dec_kernel_size': 31, 'init_type': 'xavier_uniform', 'transformer_enc_dropout_rate': 0.2, 'transformer_enc_positional_dropout_rate': 0.2, 'transformer_enc_attn_dropout_rate': 0.2, 'transformer_dec_dropout_rate': 0.2, 'transformer_dec_positional_dropout_rate': 0.2, 'transformer_dec_attn_dropout_rate': 0.2, 'pitch_predictor_layers': 5, 'pitch_predictor_chans': 256, 'pitch_predictor_kernel_size': 5, 'pitch_predictor_dropout': 0.5, 'pitch_embed_kernel_size': 1, 'pitch_embed_dropout': 0.0, 'stop_gradient_from_pitch_predictor': True, 'energy_predictor_layers': 2, 'energy_predictor_chans': 256, 'energy_predictor_kernel_size': 3, 'energy_predictor_dropout': 0.5, 'energy_embed_kernel_size': 1, 'energy_embed_dropout': 0.0, 'stop_gradient_from_energy_predictor': False, 'generator_out_channels': 1, 'generator_channels': 512, 'generator_global_channels': -1, 'generator_kernel_size': 7, 'generator_upsample_scales': [8, 8, 2, 2], 'generator_upsample_kernel_sizes': [16, 16, 4, 4], 'generator_resblock_kernel_sizes': [3, 7, 11], 'generator_resblock_dilations': [[1, 3, 5], [1, 3, 5], [1, 3, 5]], 'generator_use_additional_convs': True, 'generator_bias': True, 'generator_nonlinear_activation': 'LeakyReLU', 'generator_nonlinear_activation_params': {'negative_slope': 0.1}, 'generator_use_weight_norm': True, 'segment_size': 64, 'idim': 41, 'odim': 80}, 'discriminator_type': 'hifigan_multi_scale_multi_period_discriminator', 'discriminator_params': {'scales': 1, 'scale_downsample_pooling': 'AvgPool1d', 'scale_downsample_pooling_params': {'kernel_size': 4, 'stride': 2, 'padding': 2}, 'scale_discriminator_params': {'in_channels': 1, 'out_channels': 1, 'kernel_sizes': [15, 41, 5, 3], 'channels': 128, 'max_downsample_channels': 1024, 'max_groups': 16, 'bias': True, 'downsample_scales': [2, 2, 4, 4, 1], 'nonlinear_activation': 'LeakyReLU', 'nonlinear_activation_params': {'negative_slope': 0.1}, 'use_weight_norm': True, 'use_spectral_norm': False}, 'follow_official_norm': False, 'periods': [2, 3, 5, 7, 11], 'period_discriminator_params': {'in_channels': 1, 'out_channels': 1, 'kernel_sizes': [5, 3], 'channels': 32, 'downsample_scales': [3, 3, 3, 3, 1], 'max_downsample_channels': 1024, 'bias': True, 'nonlinear_activation': 'LeakyReLU', 'nonlinear_activation_params': {'negative_slope': 0.1}, 'use_weight_norm': True, 'use_spectral_norm': False}}, 'generator_adv_loss_params': {'average_by_discriminators': False, 'loss_type': 'mse'}, 'discriminator_adv_loss_params': {'average_by_discriminators': False, 'loss_type': 'mse'}, 'feat_match_loss_params': {'average_by_discriminators': False, 'average_by_layers': False, 'include_final_outputs': True}, 'mel_loss_params': {'fs': 24000, 'n_fft': 1024, 'hop_length': 256, 'win_length': None, 'window': 'hann', 'n_mels': 80, 'fmin': 0, 'fmax': None, 'log_base': None}, 'lambda_adv': 1.0, 'lambda_mel': 45.0, 'lambda_feat_match': 2.0, 'lambda_var': 1.0, 'lambda_align': 2.0, 'sampling_rate': 24000, 'cache_generator_outputs': True}, unused_parameters=True, use_amp=False, use_matplotlib=True, use_preprocessor=True, use_tensorboard=True, use_wandb=False, val_scheduler_criterion=('valid', 'loss'), valid_batch_bins=None, valid_batch_size=None, valid_batch_type=None, valid_data_path_and_name_and_type=[('dump/raw/jvs010_dev/text', 'text', 'text'), ('dump/raw/jvs010_dev/wav.scp', 'speech', 'sound')], valid_max_cache_size=None, valid_shape_file=['exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/valid.1.scp'], version='202204', wandb_entity=None, wandb_id=None, wandb_model_log_interval=-1, wandb_name=None, wandb_project=None, write_collected_feats=False)
+/work/espnet/espnet2/layers/stft.py:166: UserWarning: __floordiv__ is deprecated, and its behavior will change in a future version of pytorch. It currently rounds toward 0 (like the 'trunc' function NOT 'floor'). This results in incorrect rounding for negative values. To keep the current behavior, use torch.div(a, b, rounding_mode='trunc'), or for actual floor division, use torch.div(a, b, rounding_mode='floor').
+  olens = (ilens - self.n_fft) // self.hop_length + 1
+# Accounting: time=11 threads=1
+# Ended (code 0) at Tue Mar  4 21:23:37 JST 2025, elapsed time 11 seconds

exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/stats.1/config.yaml

@@ -0,0 +1,386 @@
+config: conf/tuning/train_jets.yaml
+print_config: false
+log_level: INFO
+dry_run: false
+iterator_type: sequence
+output_dir: exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/stats.1
+ngpu: 0
+seed: 777
+num_workers: 16
+num_att_plot: 3
+dist_backend: nccl
+dist_init_method: env://
+dist_world_size: null
+dist_rank: null
+local_rank: null
+dist_master_addr: null
+dist_master_port: null
+dist_launcher: null
+multiprocessing_distributed: false
+unused_parameters: true
+sharded_ddp: false
+cudnn_enabled: true
+cudnn_benchmark: false
+cudnn_deterministic: false
+collect_stats: true
+write_collected_feats: false
+max_epoch: 130
+patience: null
+val_scheduler_criterion:
+- valid
+- loss
+early_stopping_criterion:
+- valid
+- loss
+- min
+best_model_criterion:
+-   - valid
+    - text2mel_loss
+    - min
+-   - train
+    - text2mel_loss
+    - min
+-   - train
+    - total_count
+    - max
+keep_nbest_models: -1
+nbest_averaging_interval: 0
+grad_clip: -1
+grad_clip_type: 2.0
+grad_noise: false
+accum_grad: 1
+no_forward_run: false
+resume: false
+train_dtype: float32
+use_amp: false
+log_interval: 50
+use_matplotlib: true
+use_tensorboard: true
+use_wandb: false
+wandb_project: null
+wandb_id: null
+wandb_entity: null
+wandb_name: null
+wandb_model_log_interval: -1
+detect_anomaly: false
+pretrain_path: null
+init_param: []
+ignore_init_mismatch: false
+freeze_param: []
+num_iters_per_epoch: 1000
+batch_size: 20
+valid_batch_size: null
+batch_bins: 6000000
+valid_batch_bins: null
+train_shape_file:
+- exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/train.1.scp
+valid_shape_file:
+- exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/valid.1.scp
+batch_type: numel
+valid_batch_type: null
+fold_length: []
+sort_in_batch: descending
+sort_batch: descending
+multiple_iterator: false
+chunk_length: 500
+chunk_shift_ratio: 0.5
+num_cache_chunks: 1024
+train_data_path_and_name_and_type:
+-   - dump/raw/jvs010_tr_no_dev/text
+    - text
+    - text
+-   - dump/raw/jvs010_tr_no_dev/wav.scp
+    - speech
+    - sound
+valid_data_path_and_name_and_type:
+-   - dump/raw/jvs010_dev/text
+    - text
+    - text
+-   - dump/raw/jvs010_dev/wav.scp
+    - speech
+    - sound
+allow_variable_data_keys: false
+max_cache_size: 0.0
+max_cache_fd: 32
+valid_max_cache_size: null
+optim: adamw
+optim_conf:
+    lr: 0.0002
+    betas:
+    - 0.8
+    - 0.99
+    eps: 1.0e-09
+    weight_decay: 0.0
+scheduler: exponentiallr
+scheduler_conf:
+    gamma: 0.999875
+optim2: adamw
+optim2_conf:
+    lr: 0.0002
+    betas:
+    - 0.8
+    - 0.99
+    eps: 1.0e-09
+    weight_decay: 0.0
+scheduler2: exponentiallr
+scheduler2_conf:
+    gamma: 0.999875
+generator_first: true
+token_list:
+- <blank>
+- <unk>
+- o
+- a
+- u
+- i
+- e
+- k
+- r
+- t
+- n
+- pau
+- N
+- s
+- sh
+- d
+- m
+- g
+- w
+- b
+- cl
+- I
+- j
+- ch
+- y
+- U
+- h
+- p
+- ts
+- f
+- z
+- ky
+- ny
+- gy
+- ry
+- hy
+- my
+- by
+- py
+- v
+- <sos/eos>
+odim: null
+model_conf: {}
+use_preprocessor: true
+token_type: phn
+bpemodel: null
+non_linguistic_symbols: null
+cleaner: jaconv
+g2p: pyopenjtalk
+feats_extract: fbank
+feats_extract_conf:
+    n_fft: 2048
+    hop_length: 300
+    win_length: 1200
+    fs: 24000
+    fmin: 80
+    fmax: 7600
+    n_mels: 80
+normalize: null
+normalize_conf: {}
+tts: jets
+tts_conf:
+    generator_type: jets_generator
+    generator_params:
+        adim: 256
+        aheads: 2
+        elayers: 4
+        eunits: 1024
+        dlayers: 4
+        dunits: 1024
+        positionwise_layer_type: conv1d
+        positionwise_conv_kernel_size: 3
+        duration_predictor_layers: 2
+        duration_predictor_chans: 256
+        duration_predictor_kernel_size: 3
+        use_masking: true
+        encoder_normalize_before: true
+        decoder_normalize_before: true
+        encoder_type: transformer
+        decoder_type: transformer
+        conformer_rel_pos_type: latest
+        conformer_pos_enc_layer_type: rel_pos
+        conformer_self_attn_layer_type: rel_selfattn
+        conformer_activation_type: swish
+        use_macaron_style_in_conformer: true
+        use_cnn_in_conformer: true
+        conformer_enc_kernel_size: 7
+        conformer_dec_kernel_size: 31
+        init_type: xavier_uniform
+        transformer_enc_dropout_rate: 0.2
+        transformer_enc_positional_dropout_rate: 0.2
+        transformer_enc_attn_dropout_rate: 0.2
+        transformer_dec_dropout_rate: 0.2
+        transformer_dec_positional_dropout_rate: 0.2
+        transformer_dec_attn_dropout_rate: 0.2
+        pitch_predictor_layers: 5
+        pitch_predictor_chans: 256
+        pitch_predictor_kernel_size: 5
+        pitch_predictor_dropout: 0.5
+        pitch_embed_kernel_size: 1
+        pitch_embed_dropout: 0.0
+        stop_gradient_from_pitch_predictor: true
+        energy_predictor_layers: 2
+        energy_predictor_chans: 256
+        energy_predictor_kernel_size: 3
+        energy_predictor_dropout: 0.5
+        energy_embed_kernel_size: 1
+        energy_embed_dropout: 0.0
+        stop_gradient_from_energy_predictor: false
+        generator_out_channels: 1
+        generator_channels: 512
+        generator_global_channels: -1
+        generator_kernel_size: 7
+        generator_upsample_scales:
+        - 8
+        - 8
+        - 2
+        - 2
+        generator_upsample_kernel_sizes:
+        - 16
+        - 16
+        - 4
+        - 4
+        generator_resblock_kernel_sizes:
+        - 3
+        - 7
+        - 11
+        generator_resblock_dilations:
+        -   - 1
+            - 3
+            - 5
+        -   - 1
+            - 3
+            - 5
+        -   - 1
+            - 3
+            - 5
+        generator_use_additional_convs: true
+        generator_bias: true
+        generator_nonlinear_activation: LeakyReLU
+        generator_nonlinear_activation_params:
+            negative_slope: 0.1
+        generator_use_weight_norm: true
+        segment_size: 64
+        idim: 41
+        odim: 80
+    discriminator_type: hifigan_multi_scale_multi_period_discriminator
+    discriminator_params:
+        scales: 1
+        scale_downsample_pooling: AvgPool1d
+        scale_downsample_pooling_params:
+            kernel_size: 4
+            stride: 2
+            padding: 2
+        scale_discriminator_params:
+            in_channels: 1
+            out_channels: 1
+            kernel_sizes:
+            - 15
+            - 41
+            - 5
+            - 3
+            channels: 128
+            max_downsample_channels: 1024
+            max_groups: 16
+            bias: true
+            downsample_scales:
+            - 2
+            - 2
+            - 4
+            - 4
+            - 1
+            nonlinear_activation: LeakyReLU
+            nonlinear_activation_params:
+                negative_slope: 0.1
+            use_weight_norm: true
+            use_spectral_norm: false
+        follow_official_norm: false
+        periods:
+        - 2
+        - 3
+        - 5
+        - 7
+        - 11
+        period_discriminator_params:
+            in_channels: 1
+            out_channels: 1
+            kernel_sizes:
+            - 5
+            - 3
+            channels: 32
+            downsample_scales:
+            - 3
+            - 3
+            - 3
+            - 3
+            - 1
+            max_downsample_channels: 1024
+            bias: true
+            nonlinear_activation: LeakyReLU
+            nonlinear_activation_params:
+                negative_slope: 0.1
+            use_weight_norm: true
+            use_spectral_norm: false
+    generator_adv_loss_params:
+        average_by_discriminators: false
+        loss_type: mse
+    discriminator_adv_loss_params:
+        average_by_discriminators: false
+        loss_type: mse
+    feat_match_loss_params:
+        average_by_discriminators: false
+        average_by_layers: false
+        include_final_outputs: true
+    mel_loss_params:
+        fs: 24000
+        n_fft: 1024
+        hop_length: 256
+        win_length: null
+        window: hann
+        n_mels: 80
+        fmin: 0
+        fmax: null
+        log_base: null
+    lambda_adv: 1.0
+    lambda_mel: 45.0
+    lambda_feat_match: 2.0
+    lambda_var: 1.0
+    lambda_align: 2.0
+    sampling_rate: 24000
+    cache_generator_outputs: true
+pitch_extract: dio
+pitch_extract_conf:
+    reduction_factor: 1
+    use_token_averaged_f0: false
+    fs: 24000
+    n_fft: 2048
+    hop_length: 300
+    f0max: 400
+    f0min: 80
+pitch_normalize: null
+pitch_normalize_conf: {}
+energy_extract: energy
+energy_extract_conf:
+    reduction_factor: 1
+    use_token_averaged_energy: false
+    fs: 24000
+    n_fft: 2048
+    hop_length: 300
+    win_length: 1200
+energy_normalize: null
+energy_normalize_conf: {}
+required:
+- output_dir
+- token_list
+version: '202204'
+distributed: false

exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/stats.1/train/batch_keys

@@ -0,0 +1,2 @@
+text
+speech

exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/stats.1/train/energy_lengths_stats.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:71835ae88e42963ae5e3de88a9c34a4e434c2dd2662455b4e61f347e2bd7f6f2
+size 778

exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/stats.1/train/energy_stats.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5541947d077b0041a3487937ee9145a25d632bc5bbd43711f977b48ac2569c6b
+size 770

exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/stats.1/train/feats_lengths_stats.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:71835ae88e42963ae5e3de88a9c34a4e434c2dd2662455b4e61f347e2bd7f6f2
+size 778

exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/stats.1/train/feats_stats.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc5a955096b7b154b0e20365fd4a1cae7cc6fdf0e5cc0181abc8ca0dbba542af
+size 1402

exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/stats.1/train/pitch_lengths_stats.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:71835ae88e42963ae5e3de88a9c34a4e434c2dd2662455b4e61f347e2bd7f6f2
+size 778

exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/stats.1/train/pitch_stats.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:802e7de46e59805b1ba5263f3208ccfc91ab4cd6c1ca757b7ec507f8f3d07bcc
+size 770

exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/stats.1/train/speech_shape

@@ -0,0 +1,13 @@
+jvs010_VOICEACTRESS100_001 178800
+jvs010_VOICEACTRESS100_002 198240
+jvs010_VOICEACTRESS100_003 128640
+jvs010_VOICEACTRESS100_004 132000
+jvs010_VOICEACTRESS100_005 277440
+jvs010_VOICEACTRESS100_006 94560
+jvs010_VOICEACTRESS100_007 182160
+jvs010_VOICEACTRESS100_008 180960
+jvs010_VOICEACTRESS100_009 145920
+jvs010_VOICEACTRESS100_010 105359
+jvs010_VOICEACTRESS100_011 148080
+jvs010_VOICEACTRESS100_012 130320
+jvs010_VOICEACTRESS100_013 117839

exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/stats.1/train/stats_keys

@@ -0,0 +1,6 @@
+feats
+feats_lengths
+pitch
+pitch_lengths
+energy
+energy_lengths

exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/stats.1/train/text_shape

@@ -0,0 +1,13 @@
+jvs010_VOICEACTRESS100_001 78
+jvs010_VOICEACTRESS100_002 91
+jvs010_VOICEACTRESS100_003 69
+jvs010_VOICEACTRESS100_004 64
+jvs010_VOICEACTRESS100_005 121
+jvs010_VOICEACTRESS100_006 49
+jvs010_VOICEACTRESS100_007 93
+jvs010_VOICEACTRESS100_008 82
+jvs010_VOICEACTRESS100_009 69
+jvs010_VOICEACTRESS100_010 49
+jvs010_VOICEACTRESS100_011 77
+jvs010_VOICEACTRESS100_012 55
+jvs010_VOICEACTRESS100_013 54

exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/stats.1/valid/batch_keys

@@ -0,0 +1,2 @@
+text
+speech

exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/stats.1/valid/energy_lengths_stats.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:149732b1e4eb869bc4896e15295f8daa910f81fa02a9f50c1821473ebedc9e3f
+size 778

exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/stats.1/valid/energy_stats.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4931a92d9fcb8b8b598841c3589ac0e11b838be56304fd37301a3fc02a30bff5
+size 770

exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/stats.1/valid/feats_lengths_stats.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:149732b1e4eb869bc4896e15295f8daa910f81fa02a9f50c1821473ebedc9e3f
+size 778

exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/stats.1/valid/feats_stats.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dc879120415bfea6b9b8b9559285f9b75cc8881682d326559b920626c778332e
+size 1402

exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/stats.1/valid/pitch_lengths_stats.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:149732b1e4eb869bc4896e15295f8daa910f81fa02a9f50c1821473ebedc9e3f
+size 778

exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/stats.1/valid/pitch_stats.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0e277313a3d955098f9f250192f18bce2b31d2a442dc9e8a3ac909b9238e66c1
+size 770

exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/stats.1/valid/speech_shape

@@ -0,0 +1,2 @@
+jvs010_BASIC5000_0113 143040
+jvs010_BASIC5000_0261 77520

exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/stats.1/valid/stats_keys

@@ -0,0 +1,6 @@
+feats
+feats_lengths
+pitch
+pitch_lengths
+energy
+energy_lengths

exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/stats.1/valid/text_shape

@@ -0,0 +1,2 @@
+jvs010_BASIC5000_0113 56
+jvs010_BASIC5000_0261 39

exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/stats.2.log

@@ -0,0 +1,874 @@
+# python3 -m espnet2.bin.gan_tts_train --collect_stats true --write_collected_feats false --use_preprocessor true --token_type phn --token_list dump/token_list/phn_jaconv_pyopenjtalk/tokens.txt --non_linguistic_symbols none --cleaner jaconv --g2p pyopenjtalk --normalize none --pitch_normalize none --energy_normalize none --train_data_path_and_name_and_type dump/raw/jvs010_tr_no_dev/text,text,text --train_data_path_and_name_and_type dump/raw/jvs010_tr_no_dev/wav.scp,speech,sound --valid_data_path_and_name_and_type dump/raw/jvs010_dev/text,text,text --valid_data_path_and_name_and_type dump/raw/jvs010_dev/wav.scp,speech,sound --train_shape_file exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/train.2.scp --valid_shape_file exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/valid.2.scp --output_dir exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/stats.2 --config conf/tuning/train_jets.yaml --feats_extract fbank --feats_extract_conf n_fft=2048 --feats_extract_conf hop_length=300 --feats_extract_conf win_length=1200 --feats_extract_conf fs=24000 --feats_extract_conf fmin=80 --feats_extract_conf fmax=7600 --feats_extract_conf n_mels=80 --pitch_extract_conf fs=24000 --pitch_extract_conf n_fft=2048 --pitch_extract_conf hop_length=300 --pitch_extract_conf f0max=400 --pitch_extract_conf f0min=80 --energy_extract_conf fs=24000 --energy_extract_conf n_fft=2048 --energy_extract_conf hop_length=300 --energy_extract_conf win_length=1200 
+# Started at Tue Mar  4 21:23:26 JST 2025
+#
+/usr/lib/python3/dist-packages/requests/__init__.py:89: RequestsDependencyWarning: urllib3 (2.2.3) or chardet (3.0.4) doesn't match a supported version!
+  warnings.warn("urllib3 ({}) or chardet ({}) doesn't match a supported "
+/usr/bin/python3 /work/espnet/espnet2/bin/gan_tts_train.py --collect_stats true --write_collected_feats false --use_preprocessor true --token_type phn --token_list dump/token_list/phn_jaconv_pyopenjtalk/tokens.txt --non_linguistic_symbols none --cleaner jaconv --g2p pyopenjtalk --normalize none --pitch_normalize none --energy_normalize none --train_data_path_and_name_and_type dump/raw/jvs010_tr_no_dev/text,text,text --train_data_path_and_name_and_type dump/raw/jvs010_tr_no_dev/wav.scp,speech,sound --valid_data_path_and_name_and_type dump/raw/jvs010_dev/text,text,text --valid_data_path_and_name_and_type dump/raw/jvs010_dev/wav.scp,speech,sound --train_shape_file exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/train.2.scp --valid_shape_file exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/valid.2.scp --output_dir exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/stats.2 --config conf/tuning/train_jets.yaml --feats_extract fbank --feats_extract_conf n_fft=2048 --feats_extract_conf hop_length=300 --feats_extract_conf win_length=1200 --feats_extract_conf fs=24000 --feats_extract_conf fmin=80 --feats_extract_conf fmax=7600 --feats_extract_conf n_mels=80 --pitch_extract_conf fs=24000 --pitch_extract_conf n_fft=2048 --pitch_extract_conf hop_length=300 --pitch_extract_conf f0max=400 --pitch_extract_conf f0min=80 --energy_extract_conf fs=24000 --energy_extract_conf n_fft=2048 --energy_extract_conf hop_length=300 --energy_extract_conf win_length=1200
+[92b100c97f43] 2025-03-04 21:23:29,163 (gan_tts:304) INFO: Vocabulary size: 41
+[92b100c97f43] 2025-03-04 21:23:29,385 (encoder:172) INFO: encoder self-attention layer type = self-attention
+[92b100c97f43] 2025-03-04 21:23:29,507 (encoder:172) INFO: encoder self-attention layer type = self-attention
+[92b100c97f43] 2025-03-04 21:23:31,643 (abs_task:1157) INFO: pytorch.version=1.10.1+cu113, cuda.available=True, cudnn.version=8200, cudnn.benchmark=False, cudnn.deterministic=False
+[92b100c97f43] 2025-03-04 21:23:31,653 (abs_task:1158) INFO: Model structure:
+ESPnetGANTTSModel(
+  (feats_extract): LogMelFbank(
+    (stft): Stft(n_fft=2048, win_length=1200, hop_length=300, center=True, normalized=False, onesided=True)
+    (logmel): LogMel(sr=24000, n_fft=2048, n_mels=80, fmin=80, fmax=7600, htk=False)
+  )
+  (pitch_extract): Dio()
+  (energy_extract): Energy(
+    (stft): Stft(n_fft=2048, win_length=1200, hop_length=300, center=True, normalized=False, onesided=True)
+  )
+  (tts): JETS(
+    (generator): JETSGenerator(
+      (encoder): Encoder(
+        (embed): Sequential(
+          (0): Embedding(41, 256, padding_idx=0)
+          (1): ScaledPositionalEncoding(
+            (dropout): Dropout(p=0.2, inplace=False)
+          )
+        )
+        (encoders): MultiSequential(
+          (0): EncoderLayer(
+            (self_attn): MultiHeadedAttention(
+              (linear_q): Linear(in_features=256, out_features=256, bias=True)
+              (linear_k): Linear(in_features=256, out_features=256, bias=True)
+              (linear_v): Linear(in_features=256, out_features=256, bias=True)
+              (linear_out): Linear(in_features=256, out_features=256, bias=True)
+              (dropout): Dropout(p=0.2, inplace=False)
+            )
+            (feed_forward): MultiLayeredConv1d(
+              (w_1): Conv1d(256, 1024, kernel_size=(3,), stride=(1,), padding=(1,))
+              (w_2): Conv1d(1024, 256, kernel_size=(3,), stride=(1,), padding=(1,))
+              (dropout): Dropout(p=0.2, inplace=False)
+            )
+            (norm1): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (norm2): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (dropout): Dropout(p=0.2, inplace=False)
+          )
+          (1): EncoderLayer(
+            (self_attn): MultiHeadedAttention(
+              (linear_q): Linear(in_features=256, out_features=256, bias=True)
+              (linear_k): Linear(in_features=256, out_features=256, bias=True)
+              (linear_v): Linear(in_features=256, out_features=256, bias=True)
+              (linear_out): Linear(in_features=256, out_features=256, bias=True)
+              (dropout): Dropout(p=0.2, inplace=False)
+            )
+            (feed_forward): MultiLayeredConv1d(
+              (w_1): Conv1d(256, 1024, kernel_size=(3,), stride=(1,), padding=(1,))
+              (w_2): Conv1d(1024, 256, kernel_size=(3,), stride=(1,), padding=(1,))
+              (dropout): Dropout(p=0.2, inplace=False)
+            )
+            (norm1): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (norm2): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (dropout): Dropout(p=0.2, inplace=False)
+          )
+          (2): EncoderLayer(
+            (self_attn): MultiHeadedAttention(
+              (linear_q): Linear(in_features=256, out_features=256, bias=True)
+              (linear_k): Linear(in_features=256, out_features=256, bias=True)
+              (linear_v): Linear(in_features=256, out_features=256, bias=True)
+              (linear_out): Linear(in_features=256, out_features=256, bias=True)
+              (dropout): Dropout(p=0.2, inplace=False)
+            )
+            (feed_forward): MultiLayeredConv1d(
+              (w_1): Conv1d(256, 1024, kernel_size=(3,), stride=(1,), padding=(1,))
+              (w_2): Conv1d(1024, 256, kernel_size=(3,), stride=(1,), padding=(1,))
+              (dropout): Dropout(p=0.2, inplace=False)
+            )
+            (norm1): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (norm2): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (dropout): Dropout(p=0.2, inplace=False)
+          )
+          (3): EncoderLayer(
+            (self_attn): MultiHeadedAttention(
+              (linear_q): Linear(in_features=256, out_features=256, bias=True)
+              (linear_k): Linear(in_features=256, out_features=256, bias=True)
+              (linear_v): Linear(in_features=256, out_features=256, bias=True)
+              (linear_out): Linear(in_features=256, out_features=256, bias=True)
+              (dropout): Dropout(p=0.2, inplace=False)
+            )
+            (feed_forward): MultiLayeredConv1d(
+              (w_1): Conv1d(256, 1024, kernel_size=(3,), stride=(1,), padding=(1,))
+              (w_2): Conv1d(1024, 256, kernel_size=(3,), stride=(1,), padding=(1,))
+              (dropout): Dropout(p=0.2, inplace=False)
+            )
+            (norm1): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (norm2): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (dropout): Dropout(p=0.2, inplace=False)
+          )
+        )
+        (after_norm): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+      )
+      (duration_predictor): DurationPredictor(
+        (conv): ModuleList(
+          (0): Sequential(
+            (0): Conv1d(256, 256, kernel_size=(3,), stride=(1,), padding=(1,))
+            (1): ReLU()
+            (2): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (3): Dropout(p=0.1, inplace=False)
+          )
+          (1): Sequential(
+            (0): Conv1d(256, 256, kernel_size=(3,), stride=(1,), padding=(1,))
+            (1): ReLU()
+            (2): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (3): Dropout(p=0.1, inplace=False)
+          )
+        )
+        (linear): Linear(in_features=256, out_features=1, bias=True)
+      )
+      (pitch_predictor): VariancePredictor(
+        (conv): ModuleList(
+          (0): Sequential(
+            (0): Conv1d(256, 256, kernel_size=(5,), stride=(1,), padding=(2,))
+            (1): ReLU()
+            (2): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (3): Dropout(p=0.5, inplace=False)
+          )
+          (1): Sequential(
+            (0): Conv1d(256, 256, kernel_size=(5,), stride=(1,), padding=(2,))
+            (1): ReLU()
+            (2): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (3): Dropout(p=0.5, inplace=False)
+          )
+          (2): Sequential(
+            (0): Conv1d(256, 256, kernel_size=(5,), stride=(1,), padding=(2,))
+            (1): ReLU()
+            (2): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (3): Dropout(p=0.5, inplace=False)
+          )
+          (3): Sequential(
+            (0): Conv1d(256, 256, kernel_size=(5,), stride=(1,), padding=(2,))
+            (1): ReLU()
+            (2): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (3): Dropout(p=0.5, inplace=False)
+          )
+          (4): Sequential(
+            (0): Conv1d(256, 256, kernel_size=(5,), stride=(1,), padding=(2,))
+            (1): ReLU()
+            (2): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (3): Dropout(p=0.5, inplace=False)
+          )
+        )
+        (linear): Linear(in_features=256, out_features=1, bias=True)
+      )
+      (pitch_embed): Sequential(
+        (0): Conv1d(1, 256, kernel_size=(1,), stride=(1,))
+        (1): Dropout(p=0.0, inplace=False)
+      )
+      (energy_predictor): VariancePredictor(
+        (conv): ModuleList(
+          (0): Sequential(
+            (0): Conv1d(256, 256, kernel_size=(3,), stride=(1,), padding=(1,))
+            (1): ReLU()
+            (2): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (3): Dropout(p=0.5, inplace=False)
+          )
+          (1): Sequential(
+            (0): Conv1d(256, 256, kernel_size=(3,), stride=(1,), padding=(1,))
+            (1): ReLU()
+            (2): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (3): Dropout(p=0.5, inplace=False)
+          )
+        )
+        (linear): Linear(in_features=256, out_features=1, bias=True)
+      )
+      (energy_embed): Sequential(
+        (0): Conv1d(1, 256, kernel_size=(1,), stride=(1,))
+        (1): Dropout(p=0.0, inplace=False)
+      )
+      (alignment_module): AlignmentModule(
+        (t_conv1): Conv1d(256, 256, kernel_size=(3,), stride=(1,), padding=(1,))
+        (t_conv2): Conv1d(256, 256, kernel_size=(1,), stride=(1,))
+        (f_conv1): Conv1d(80, 256, kernel_size=(3,), stride=(1,), padding=(1,))
+        (f_conv2): Conv1d(256, 256, kernel_size=(3,), stride=(1,), padding=(1,))
+        (f_conv3): Conv1d(256, 256, kernel_size=(1,), stride=(1,))
+      )
+      (length_regulator): GaussianUpsampling()
+      (decoder): Encoder(
+        (embed): Sequential(
+          (0): ScaledPositionalEncoding(
+            (dropout): Dropout(p=0.2, inplace=False)
+          )
+        )
+        (encoders): MultiSequential(
+          (0): EncoderLayer(
+            (self_attn): MultiHeadedAttention(
+              (linear_q): Linear(in_features=256, out_features=256, bias=True)
+              (linear_k): Linear(in_features=256, out_features=256, bias=True)
+              (linear_v): Linear(in_features=256, out_features=256, bias=True)
+              (linear_out): Linear(in_features=256, out_features=256, bias=True)
+              (dropout): Dropout(p=0.2, inplace=False)
+            )
+            (feed_forward): MultiLayeredConv1d(
+              (w_1): Conv1d(256, 1024, kernel_size=(3,), stride=(1,), padding=(1,))
+              (w_2): Conv1d(1024, 256, kernel_size=(3,), stride=(1,), padding=(1,))
+              (dropout): Dropout(p=0.2, inplace=False)
+            )
+            (norm1): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (norm2): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (dropout): Dropout(p=0.2, inplace=False)
+          )
+          (1): EncoderLayer(
+            (self_attn): MultiHeadedAttention(
+              (linear_q): Linear(in_features=256, out_features=256, bias=True)
+              (linear_k): Linear(in_features=256, out_features=256, bias=True)
+              (linear_v): Linear(in_features=256, out_features=256, bias=True)
+              (linear_out): Linear(in_features=256, out_features=256, bias=True)
+              (dropout): Dropout(p=0.2, inplace=False)
+            )
+            (feed_forward): MultiLayeredConv1d(
+              (w_1): Conv1d(256, 1024, kernel_size=(3,), stride=(1,), padding=(1,))
+              (w_2): Conv1d(1024, 256, kernel_size=(3,), stride=(1,), padding=(1,))
+              (dropout): Dropout(p=0.2, inplace=False)
+            )
+            (norm1): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (norm2): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (dropout): Dropout(p=0.2, inplace=False)
+          )
+          (2): EncoderLayer(
+            (self_attn): MultiHeadedAttention(
+              (linear_q): Linear(in_features=256, out_features=256, bias=True)
+              (linear_k): Linear(in_features=256, out_features=256, bias=True)
+              (linear_v): Linear(in_features=256, out_features=256, bias=True)
+              (linear_out): Linear(in_features=256, out_features=256, bias=True)
+              (dropout): Dropout(p=0.2, inplace=False)
+            )
+            (feed_forward): MultiLayeredConv1d(
+              (w_1): Conv1d(256, 1024, kernel_size=(3,), stride=(1,), padding=(1,))
+              (w_2): Conv1d(1024, 256, kernel_size=(3,), stride=(1,), padding=(1,))
+              (dropout): Dropout(p=0.2, inplace=False)
+            )
+            (norm1): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (norm2): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (dropout): Dropout(p=0.2, inplace=False)
+          )
+          (3): EncoderLayer(
+            (self_attn): MultiHeadedAttention(
+              (linear_q): Linear(in_features=256, out_features=256, bias=True)
+              (linear_k): Linear(in_features=256, out_features=256, bias=True)
+              (linear_v): Linear(in_features=256, out_features=256, bias=True)
+              (linear_out): Linear(in_features=256, out_features=256, bias=True)
+              (dropout): Dropout(p=0.2, inplace=False)
+            )
+            (feed_forward): MultiLayeredConv1d(
+              (w_1): Conv1d(256, 1024, kernel_size=(3,), stride=(1,), padding=(1,))
+              (w_2): Conv1d(1024, 256, kernel_size=(3,), stride=(1,), padding=(1,))
+              (dropout): Dropout(p=0.2, inplace=False)
+            )
+            (norm1): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (norm2): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (dropout): Dropout(p=0.2, inplace=False)
+          )
+        )
+        (after_norm): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+      )
+      (generator): HiFiGANGenerator(
+        (input_conv): Conv1d(256, 512, kernel_size=(7,), stride=(1,), padding=(3,))
+        (upsamples): ModuleList(
+          (0): Sequential(
+            (0): LeakyReLU(negative_slope=0.1)
+            (1): ConvTranspose1d(512, 256, kernel_size=(16,), stride=(8,), padding=(4,))
+          )
+          (1): Sequential(
+            (0): LeakyReLU(negative_slope=0.1)
+            (1): ConvTranspose1d(256, 128, kernel_size=(16,), stride=(8,), padding=(4,))
+          )
+          (2): Sequential(
+            (0): LeakyReLU(negative_slope=0.1)
+            (1): ConvTranspose1d(128, 64, kernel_size=(4,), stride=(2,), padding=(1,))
+          )
+          (3): Sequential(
+            (0): LeakyReLU(negative_slope=0.1)
+            (1): ConvTranspose1d(64, 32, kernel_size=(4,), stride=(2,), padding=(1,))
+          )
+        )
+        (blocks): ModuleList(
+          (0): ResidualBlock(
+            (convs1): ModuleList(
+              (0): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(256, 256, kernel_size=(3,), stride=(1,), padding=(1,))
+              )
+              (1): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(256, 256, kernel_size=(3,), stride=(1,), padding=(3,), dilation=(3,))
+              )
+              (2): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(256, 256, kernel_size=(3,), stride=(1,), padding=(5,), dilation=(5,))
+              )
+            )
+            (convs2): ModuleList(
+              (0): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(256, 256, kernel_size=(3,), stride=(1,), padding=(1,))
+              )
+              (1): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(256, 256, kernel_size=(3,), stride=(1,), padding=(1,))
+              )
+              (2): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(256, 256, kernel_size=(3,), stride=(1,), padding=(1,))
+              )
+            )
+          )
+          (1): ResidualBlock(
+            (convs1): ModuleList(
+              (0): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(256, 256, kernel_size=(7,), stride=(1,), padding=(3,))
+              )
+              (1): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(256, 256, kernel_size=(7,), stride=(1,), padding=(9,), dilation=(3,))
+              )
+              (2): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(256, 256, kernel_size=(7,), stride=(1,), padding=(15,), dilation=(5,))
+              )
+            )
+            (convs2): ModuleList(
+              (0): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(256, 256, kernel_size=(7,), stride=(1,), padding=(3,))
+              )
+              (1): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(256, 256, kernel_size=(7,), stride=(1,), padding=(3,))
+              )
+              (2): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(256, 256, kernel_size=(7,), stride=(1,), padding=(3,))
+              )
+            )
+          )
+          (2): ResidualBlock(
+            (convs1): ModuleList(
+              (0): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(256, 256, kernel_size=(11,), stride=(1,), padding=(5,))
+              )
+              (1): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(256, 256, kernel_size=(11,), stride=(1,), padding=(15,), dilation=(3,))
+              )
+              (2): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(256, 256, kernel_size=(11,), stride=(1,), padding=(25,), dilation=(5,))
+              )
+            )
+            (convs2): ModuleList(
+              (0): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(256, 256, kernel_size=(11,), stride=(1,), padding=(5,))
+              )
+              (1): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(256, 256, kernel_size=(11,), stride=(1,), padding=(5,))
+              )
+              (2): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(256, 256, kernel_size=(11,), stride=(1,), padding=(5,))
+              )
+            )
+          )
+          (3): ResidualBlock(
+            (convs1): ModuleList(
+              (0): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(128, 128, kernel_size=(3,), stride=(1,), padding=(1,))
+              )
+              (1): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(128, 128, kernel_size=(3,), stride=(1,), padding=(3,), dilation=(3,))
+              )
+              (2): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(128, 128, kernel_size=(3,), stride=(1,), padding=(5,), dilation=(5,))
+              )
+            )
+            (convs2): ModuleList(
+              (0): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(128, 128, kernel_size=(3,), stride=(1,), padding=(1,))
+              )
+              (1): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(128, 128, kernel_size=(3,), stride=(1,), padding=(1,))
+              )
+              (2): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(128, 128, kernel_size=(3,), stride=(1,), padding=(1,))
+              )
+            )
+          )
+          (4): ResidualBlock(
+            (convs1): ModuleList(
+              (0): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(128, 128, kernel_size=(7,), stride=(1,), padding=(3,))
+              )
+              (1): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(128, 128, kernel_size=(7,), stride=(1,), padding=(9,), dilation=(3,))
+              )
+              (2): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(128, 128, kernel_size=(7,), stride=(1,), padding=(15,), dilation=(5,))
+              )
+            )
+            (convs2): ModuleList(
+              (0): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(128, 128, kernel_size=(7,), stride=(1,), padding=(3,))
+              )
+              (1): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(128, 128, kernel_size=(7,), stride=(1,), padding=(3,))
+              )
+              (2): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(128, 128, kernel_size=(7,), stride=(1,), padding=(3,))
+              )
+            )
+          )
+          (5): ResidualBlock(
+            (convs1): ModuleList(
+              (0): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(128, 128, kernel_size=(11,), stride=(1,), padding=(5,))
+              )
+              (1): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(128, 128, kernel_size=(11,), stride=(1,), padding=(15,), dilation=(3,))
+              )
+              (2): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(128, 128, kernel_size=(11,), stride=(1,), padding=(25,), dilation=(5,))
+              )
+            )
+            (convs2): ModuleList(
+              (0): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(128, 128, kernel_size=(11,), stride=(1,), padding=(5,))
+              )
+              (1): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(128, 128, kernel_size=(11,), stride=(1,), padding=(5,))
+              )
+              (2): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(128, 128, kernel_size=(11,), stride=(1,), padding=(5,))
+              )
+            )
+          )
+          (6): ResidualBlock(
+            (convs1): ModuleList(
+              (0): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(64, 64, kernel_size=(3,), stride=(1,), padding=(1,))
+              )
+              (1): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(64, 64, kernel_size=(3,), stride=(1,), padding=(3,), dilation=(3,))
+              )
+              (2): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(64, 64, kernel_size=(3,), stride=(1,), padding=(5,), dilation=(5,))
+              )
+            )
+            (convs2): ModuleList(
+              (0): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(64, 64, kernel_size=(3,), stride=(1,), padding=(1,))
+              )
+              (1): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(64, 64, kernel_size=(3,), stride=(1,), padding=(1,))
+              )
+              (2): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(64, 64, kernel_size=(3,), stride=(1,), padding=(1,))
+              )
+            )
+          )
+          (7): ResidualBlock(
+            (convs1): ModuleList(
+              (0): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(64, 64, kernel_size=(7,), stride=(1,), padding=(3,))
+              )
+              (1): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(64, 64, kernel_size=(7,), stride=(1,), padding=(9,), dilation=(3,))
+              )
+              (2): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(64, 64, kernel_size=(7,), stride=(1,), padding=(15,), dilation=(5,))
+              )
+            )
+            (convs2): ModuleList(
+              (0): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(64, 64, kernel_size=(7,), stride=(1,), padding=(3,))
+              )
+              (1): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(64, 64, kernel_size=(7,), stride=(1,), padding=(3,))
+              )
+              (2): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(64, 64, kernel_size=(7,), stride=(1,), padding=(3,))
+              )
+            )
+          )
+          (8): ResidualBlock(
+            (convs1): ModuleList(
+              (0): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(64, 64, kernel_size=(11,), stride=(1,), padding=(5,))
+              )
+              (1): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(64, 64, kernel_size=(11,), stride=(1,), padding=(15,), dilation=(3,))
+              )
+              (2): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(64, 64, kernel_size=(11,), stride=(1,), padding=(25,), dilation=(5,))
+              )
+            )
+            (convs2): ModuleList(
+              (0): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(64, 64, kernel_size=(11,), stride=(1,), padding=(5,))
+              )
+              (1): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(64, 64, kernel_size=(11,), stride=(1,), padding=(5,))
+              )
+              (2): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(64, 64, kernel_size=(11,), stride=(1,), padding=(5,))
+              )
+            )
+          )
+          (9): ResidualBlock(
+            (convs1): ModuleList(
+              (0): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(32, 32, kernel_size=(3,), stride=(1,), padding=(1,))
+              )
+              (1): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(32, 32, kernel_size=(3,), stride=(1,), padding=(3,), dilation=(3,))
+              )
+              (2): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(32, 32, kernel_size=(3,), stride=(1,), padding=(5,), dilation=(5,))
+              )
+            )
+            (convs2): ModuleList(
+              (0): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(32, 32, kernel_size=(3,), stride=(1,), padding=(1,))
+              )
+              (1): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(32, 32, kernel_size=(3,), stride=(1,), padding=(1,))
+              )
+              (2): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(32, 32, kernel_size=(3,), stride=(1,), padding=(1,))
+              )
+            )
+          )
+          (10): ResidualBlock(
+            (convs1): ModuleList(
+              (0): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(32, 32, kernel_size=(7,), stride=(1,), padding=(3,))
+              )
+              (1): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(32, 32, kernel_size=(7,), stride=(1,), padding=(9,), dilation=(3,))
+              )
+              (2): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(32, 32, kernel_size=(7,), stride=(1,), padding=(15,), dilation=(5,))
+              )
+            )
+            (convs2): ModuleList(
+              (0): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(32, 32, kernel_size=(7,), stride=(1,), padding=(3,))
+              )
+              (1): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(32, 32, kernel_size=(7,), stride=(1,), padding=(3,))
+              )
+              (2): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(32, 32, kernel_size=(7,), stride=(1,), padding=(3,))
+              )
+            )
+          )
+          (11): ResidualBlock(
+            (convs1): ModuleList(
+              (0): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(32, 32, kernel_size=(11,), stride=(1,), padding=(5,))
+              )
+              (1): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(32, 32, kernel_size=(11,), stride=(1,), padding=(15,), dilation=(3,))
+              )
+              (2): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(32, 32, kernel_size=(11,), stride=(1,), padding=(25,), dilation=(5,))
+              )
+            )
+            (convs2): ModuleList(
+              (0): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(32, 32, kernel_size=(11,), stride=(1,), padding=(5,))
+              )
+              (1): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(32, 32, kernel_size=(11,), stride=(1,), padding=(5,))
+              )
+              (2): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(32, 32, kernel_size=(11,), stride=(1,), padding=(5,))
+              )
+            )
+          )
+        )
+        (output_conv): Sequential(
+          (0): LeakyReLU(negative_slope=0.01)
+          (1): Conv1d(32, 1, kernel_size=(7,), stride=(1,), padding=(3,))
+          (2): Tanh()
+        )
+      )
+    )
+    (discriminator): HiFiGANMultiScaleMultiPeriodDiscriminator(
+      (msd): HiFiGANMultiScaleDiscriminator(
+        (discriminators): ModuleList(
+          (0): HiFiGANScaleDiscriminator(
+            (layers): ModuleList(
+              (0): Sequential(
+                (0): Conv1d(1, 128, kernel_size=(15,), stride=(1,), padding=(7,))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (1): Sequential(
+                (0): Conv1d(128, 128, kernel_size=(41,), stride=(2,), padding=(20,), groups=4)
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (2): Sequential(
+                (0): Conv1d(128, 256, kernel_size=(41,), stride=(2,), padding=(20,), groups=16)
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (3): Sequential(
+                (0): Conv1d(256, 512, kernel_size=(41,), stride=(4,), padding=(20,), groups=16)
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (4): Sequential(
+                (0): Conv1d(512, 1024, kernel_size=(41,), stride=(4,), padding=(20,), groups=16)
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (5): Sequential(
+                (0): Conv1d(1024, 1024, kernel_size=(41,), stride=(1,), padding=(20,), groups=16)
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (6): Sequential(
+                (0): Conv1d(1024, 1024, kernel_size=(5,), stride=(1,), padding=(2,))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (7): Conv1d(1024, 1, kernel_size=(3,), stride=(1,), padding=(1,))
+            )
+          )
+        )
+      )
+      (mpd): HiFiGANMultiPeriodDiscriminator(
+        (discriminators): ModuleList(
+          (0): HiFiGANPeriodDiscriminator(
+            (convs): ModuleList(
+              (0): Sequential(
+                (0): Conv2d(1, 32, kernel_size=(5, 1), stride=(3, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (1): Sequential(
+                (0): Conv2d(32, 128, kernel_size=(5, 1), stride=(3, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (2): Sequential(
+                (0): Conv2d(128, 512, kernel_size=(5, 1), stride=(3, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (3): Sequential(
+                (0): Conv2d(512, 1024, kernel_size=(5, 1), stride=(3, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (4): Sequential(
+                (0): Conv2d(1024, 1024, kernel_size=(5, 1), stride=(1, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+            )
+            (output_conv): Conv2d(1024, 1, kernel_size=(2, 1), stride=(1, 1), padding=(1, 0))
+          )
+          (1): HiFiGANPeriodDiscriminator(
+            (convs): ModuleList(
+              (0): Sequential(
+                (0): Conv2d(1, 32, kernel_size=(5, 1), stride=(3, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (1): Sequential(
+                (0): Conv2d(32, 128, kernel_size=(5, 1), stride=(3, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (2): Sequential(
+                (0): Conv2d(128, 512, kernel_size=(5, 1), stride=(3, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (3): Sequential(
+                (0): Conv2d(512, 1024, kernel_size=(5, 1), stride=(3, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (4): Sequential(
+                (0): Conv2d(1024, 1024, kernel_size=(5, 1), stride=(1, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+            )
+            (output_conv): Conv2d(1024, 1, kernel_size=(2, 1), stride=(1, 1), padding=(1, 0))
+          )
+          (2): HiFiGANPeriodDiscriminator(
+            (convs): ModuleList(
+              (0): Sequential(
+                (0): Conv2d(1, 32, kernel_size=(5, 1), stride=(3, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (1): Sequential(
+                (0): Conv2d(32, 128, kernel_size=(5, 1), stride=(3, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (2): Sequential(
+                (0): Conv2d(128, 512, kernel_size=(5, 1), stride=(3, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (3): Sequential(
+                (0): Conv2d(512, 1024, kernel_size=(5, 1), stride=(3, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (4): Sequential(
+                (0): Conv2d(1024, 1024, kernel_size=(5, 1), stride=(1, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+            )
+            (output_conv): Conv2d(1024, 1, kernel_size=(2, 1), stride=(1, 1), padding=(1, 0))
+          )
+          (3): HiFiGANPeriodDiscriminator(
+            (convs): ModuleList(
+              (0): Sequential(
+                (0): Conv2d(1, 32, kernel_size=(5, 1), stride=(3, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (1): Sequential(
+                (0): Conv2d(32, 128, kernel_size=(5, 1), stride=(3, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (2): Sequential(
+                (0): Conv2d(128, 512, kernel_size=(5, 1), stride=(3, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (3): Sequential(
+                (0): Conv2d(512, 1024, kernel_size=(5, 1), stride=(3, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (4): Sequential(
+                (0): Conv2d(1024, 1024, kernel_size=(5, 1), stride=(1, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+            )
+            (output_conv): Conv2d(1024, 1, kernel_size=(2, 1), stride=(1, 1), padding=(1, 0))
+          )
+          (4): HiFiGANPeriodDiscriminator(
+            (convs): ModuleList(
+              (0): Sequential(
+                (0): Conv2d(1, 32, kernel_size=(5, 1), stride=(3, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (1): Sequential(
+                (0): Conv2d(32, 128, kernel_size=(5, 1), stride=(3, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (2): Sequential(
+                (0): Conv2d(128, 512, kernel_size=(5, 1), stride=(3, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (3): Sequential(
+                (0): Conv2d(512, 1024, kernel_size=(5, 1), stride=(3, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (4): Sequential(
+                (0): Conv2d(1024, 1024, kernel_size=(5, 1), stride=(1, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+            )
+            (output_conv): Conv2d(1024, 1, kernel_size=(2, 1), stride=(1, 1), padding=(1, 0))
+          )
+        )
+      )
+    )
+    (generator_adv_loss): GeneratorAdversarialLoss()
+    (discriminator_adv_loss): DiscriminatorAdversarialLoss()
+    (feat_match_loss): FeatureMatchLoss()
+    (mel_loss): MelSpectrogramLoss(
+      (wav_to_mel): LogMelFbank(
+        (stft): Stft(n_fft=1024, win_length=1024, hop_length=256, center=True, normalized=False, onesided=True)
+        (logmel): LogMel(sr=24000, n_fft=1024, n_mels=80, fmin=0, fmax=12000.0, htk=False)
+      )
+    )
+    (var_loss): VarianceLoss(
+      (mse_criterion): MSELoss()
+      (duration_criterion): DurationPredictorLoss(
+        (criterion): MSELoss()
+      )
+    )
+    (forwardsum_loss): ForwardSumLoss()
+  )
+)
+
+Model summary:
+    Class Name: ESPnetGANTTSModel
+    Total Number of model parameters: 83.28 M
+    Number of trainable parameters: 83.28 M (100.0%)
+    Size: 333.11 MB
+    Type: torch.float32
+[92b100c97f43] 2025-03-04 21:23:31,654 (abs_task:1161) INFO: Optimizer:
+AdamW (
+Parameter Group 0
+    amsgrad: False
+    betas: [0.8, 0.99]
+    eps: 1e-09
+    initial_lr: 0.0002
+    lr: 0.0002
+    weight_decay: 0.0
+)
+[92b100c97f43] 2025-03-04 21:23:31,654 (abs_task:1162) INFO: Scheduler: <torch.optim.lr_scheduler.ExponentialLR object at 0x7f31cda9c160>
+[92b100c97f43] 2025-03-04 21:23:31,654 (abs_task:1161) INFO: Optimizer2:
+AdamW (
+Parameter Group 0
+    amsgrad: False
+    betas: [0.8, 0.99]
+    eps: 1e-09
+    initial_lr: 0.0002
+    lr: 0.0002
+    weight_decay: 0.0
+)
+[92b100c97f43] 2025-03-04 21:23:31,654 (abs_task:1162) INFO: Scheduler2: <torch.optim.lr_scheduler.ExponentialLR object at 0x7f31dc9c2280>
+[92b100c97f43] 2025-03-04 21:23:31,654 (abs_task:1171) INFO: Saving the configuration in exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/stats.2/config.yaml
+[92b100c97f43] 2025-03-04 21:23:31,672 (abs_task:1182) INFO: Namespace(accum_grad=1, allow_variable_data_keys=False, batch_bins=6000000, batch_size=20, batch_type='numel', best_model_criterion=[['valid', 'text2mel_loss', 'min'], ['train', 'text2mel_loss', 'min'], ['train', 'total_count', 'max']], bpemodel=None, chunk_length=500, chunk_shift_ratio=0.5, cleaner='jaconv', collect_stats=True, config='conf/tuning/train_jets.yaml', cudnn_benchmark=False, cudnn_deterministic=False, cudnn_enabled=True, detect_anomaly=False, dist_backend='nccl', dist_init_method='env://', dist_launcher=None, dist_master_addr=None, dist_master_port=None, dist_rank=None, dist_world_size=None, distributed=False, dry_run=False, early_stopping_criterion=('valid', 'loss', 'min'), energy_extract='energy', energy_extract_conf={'reduction_factor': 1, 'use_token_averaged_energy': False, 'fs': 24000, 'n_fft': 2048, 'hop_length': 300, 'win_length': 1200}, energy_normalize=None, energy_normalize_conf={}, feats_extract='fbank', feats_extract_conf={'n_fft': 2048, 'hop_length': 300, 'win_length': 1200, 'fs': 24000, 'fmin': 80, 'fmax': 7600, 'n_mels': 80}, fold_length=[], freeze_param=[], g2p='pyopenjtalk', generator_first=True, grad_clip=-1, grad_clip_type=2.0, grad_noise=False, ignore_init_mismatch=False, init_param=[], iterator_type='sequence', keep_nbest_models=-1, local_rank=None, log_interval=50, log_level='INFO', max_cache_fd=32, max_cache_size=0.0, max_epoch=130, model_conf={}, multiple_iterator=False, multiprocessing_distributed=False, nbest_averaging_interval=0, ngpu=0, no_forward_run=False, non_linguistic_symbols=None, normalize=None, normalize_conf={}, num_att_plot=3, num_cache_chunks=1024, num_iters_per_epoch=1000, num_workers=16, odim=None, optim='adamw', optim2='adamw', optim2_conf={'lr': 0.0002, 'betas': [0.8, 0.99], 'eps': 1e-09, 'weight_decay': 0.0}, optim_conf={'lr': 0.0002, 'betas': [0.8, 0.99], 'eps': 1e-09, 'weight_decay': 0.0}, output_dir='exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/stats.2', patience=None, pitch_extract='dio', pitch_extract_conf={'reduction_factor': 1, 'use_token_averaged_f0': False, 'fs': 24000, 'n_fft': 2048, 'hop_length': 300, 'f0max': 400, 'f0min': 80}, pitch_normalize=None, pitch_normalize_conf={}, pretrain_path=None, print_config=False, required=['output_dir', 'token_list'], resume=False, scheduler='exponentiallr', scheduler2='exponentiallr', scheduler2_conf={'gamma': 0.999875}, scheduler_conf={'gamma': 0.999875}, seed=777, sharded_ddp=False, sort_batch='descending', sort_in_batch='descending', token_list=['<blank>', '<unk>', 'o', 'a', 'u', 'i', 'e', 'k', 'r', 't', 'n', 'pau', 'N', 's', 'sh', 'd', 'm', 'g', 'w', 'b', 'cl', 'I', 'j', 'ch', 'y', 'U', 'h', 'p', 'ts', 'f', 'z', 'ky', 'ny', 'gy', 'ry', 'hy', 'my', 'by', 'py', 'v', '<sos/eos>'], token_type='phn', train_data_path_and_name_and_type=[('dump/raw/jvs010_tr_no_dev/text', 'text', 'text'), ('dump/raw/jvs010_tr_no_dev/wav.scp', 'speech', 'sound')], train_dtype='float32', train_shape_file=['exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/train.2.scp'], tts='jets', tts_conf={'generator_type': 'jets_generator', 'generator_params': {'adim': 256, 'aheads': 2, 'elayers': 4, 'eunits': 1024, 'dlayers': 4, 'dunits': 1024, 'positionwise_layer_type': 'conv1d', 'positionwise_conv_kernel_size': 3, 'duration_predictor_layers': 2, 'duration_predictor_chans': 256, 'duration_predictor_kernel_size': 3, 'use_masking': True, 'encoder_normalize_before': True, 'decoder_normalize_before': True, 'encoder_type': 'transformer', 'decoder_type': 'transformer', 'conformer_rel_pos_type': 'latest', 'conformer_pos_enc_layer_type': 'rel_pos', 'conformer_self_attn_layer_type': 'rel_selfattn', 'conformer_activation_type': 'swish', 'use_macaron_style_in_conformer': True, 'use_cnn_in_conformer': True, 'conformer_enc_kernel_size': 7, 'conformer_dec_kernel_size': 31, 'init_type': 'xavier_uniform', 'transformer_enc_dropout_rate': 0.2, 'transformer_enc_positional_dropout_rate': 0.2, 'transformer_enc_attn_dropout_rate': 0.2, 'transformer_dec_dropout_rate': 0.2, 'transformer_dec_positional_dropout_rate': 0.2, 'transformer_dec_attn_dropout_rate': 0.2, 'pitch_predictor_layers': 5, 'pitch_predictor_chans': 256, 'pitch_predictor_kernel_size': 5, 'pitch_predictor_dropout': 0.5, 'pitch_embed_kernel_size': 1, 'pitch_embed_dropout': 0.0, 'stop_gradient_from_pitch_predictor': True, 'energy_predictor_layers': 2, 'energy_predictor_chans': 256, 'energy_predictor_kernel_size': 3, 'energy_predictor_dropout': 0.5, 'energy_embed_kernel_size': 1, 'energy_embed_dropout': 0.0, 'stop_gradient_from_energy_predictor': False, 'generator_out_channels': 1, 'generator_channels': 512, 'generator_global_channels': -1, 'generator_kernel_size': 7, 'generator_upsample_scales': [8, 8, 2, 2], 'generator_upsample_kernel_sizes': [16, 16, 4, 4], 'generator_resblock_kernel_sizes': [3, 7, 11], 'generator_resblock_dilations': [[1, 3, 5], [1, 3, 5], [1, 3, 5]], 'generator_use_additional_convs': True, 'generator_bias': True, 'generator_nonlinear_activation': 'LeakyReLU', 'generator_nonlinear_activation_params': {'negative_slope': 0.1}, 'generator_use_weight_norm': True, 'segment_size': 64, 'idim': 41, 'odim': 80}, 'discriminator_type': 'hifigan_multi_scale_multi_period_discriminator', 'discriminator_params': {'scales': 1, 'scale_downsample_pooling': 'AvgPool1d', 'scale_downsample_pooling_params': {'kernel_size': 4, 'stride': 2, 'padding': 2}, 'scale_discriminator_params': {'in_channels': 1, 'out_channels': 1, 'kernel_sizes': [15, 41, 5, 3], 'channels': 128, 'max_downsample_channels': 1024, 'max_groups': 16, 'bias': True, 'downsample_scales': [2, 2, 4, 4, 1], 'nonlinear_activation': 'LeakyReLU', 'nonlinear_activation_params': {'negative_slope': 0.1}, 'use_weight_norm': True, 'use_spectral_norm': False}, 'follow_official_norm': False, 'periods': [2, 3, 5, 7, 11], 'period_discriminator_params': {'in_channels': 1, 'out_channels': 1, 'kernel_sizes': [5, 3], 'channels': 32, 'downsample_scales': [3, 3, 3, 3, 1], 'max_downsample_channels': 1024, 'bias': True, 'nonlinear_activation': 'LeakyReLU', 'nonlinear_activation_params': {'negative_slope': 0.1}, 'use_weight_norm': True, 'use_spectral_norm': False}}, 'generator_adv_loss_params': {'average_by_discriminators': False, 'loss_type': 'mse'}, 'discriminator_adv_loss_params': {'average_by_discriminators': False, 'loss_type': 'mse'}, 'feat_match_loss_params': {'average_by_discriminators': False, 'average_by_layers': False, 'include_final_outputs': True}, 'mel_loss_params': {'fs': 24000, 'n_fft': 1024, 'hop_length': 256, 'win_length': None, 'window': 'hann', 'n_mels': 80, 'fmin': 0, 'fmax': None, 'log_base': None}, 'lambda_adv': 1.0, 'lambda_mel': 45.0, 'lambda_feat_match': 2.0, 'lambda_var': 1.0, 'lambda_align': 2.0, 'sampling_rate': 24000, 'cache_generator_outputs': True}, unused_parameters=True, use_amp=False, use_matplotlib=True, use_preprocessor=True, use_tensorboard=True, use_wandb=False, val_scheduler_criterion=('valid', 'loss'), valid_batch_bins=None, valid_batch_size=None, valid_batch_type=None, valid_data_path_and_name_and_type=[('dump/raw/jvs010_dev/text', 'text', 'text'), ('dump/raw/jvs010_dev/wav.scp', 'speech', 'sound')], valid_max_cache_size=None, valid_shape_file=['exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/valid.2.scp'], version='202204', wandb_entity=None, wandb_id=None, wandb_model_log_interval=-1, wandb_name=None, wandb_project=None, write_collected_feats=False)
+/work/espnet/espnet2/layers/stft.py:166: UserWarning: __floordiv__ is deprecated, and its behavior will change in a future version of pytorch. It currently rounds toward 0 (like the 'trunc' function NOT 'floor'). This results in incorrect rounding for negative values. To keep the current behavior, use torch.div(a, b, rounding_mode='trunc'), or for actual floor division, use torch.div(a, b, rounding_mode='floor').
+  olens = (ilens - self.n_fft) // self.hop_length + 1
+# Accounting: time=11 threads=1
+# Ended (code 0) at Tue Mar  4 21:23:37 JST 2025, elapsed time 11 seconds

exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/stats.2/config.yaml

@@ -0,0 +1,386 @@
+config: conf/tuning/train_jets.yaml
+print_config: false
+log_level: INFO
+dry_run: false
+iterator_type: sequence
+output_dir: exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/stats.2
+ngpu: 0
+seed: 777
+num_workers: 16
+num_att_plot: 3
+dist_backend: nccl
+dist_init_method: env://
+dist_world_size: null
+dist_rank: null
+local_rank: null
+dist_master_addr: null
+dist_master_port: null
+dist_launcher: null
+multiprocessing_distributed: false
+unused_parameters: true
+sharded_ddp: false
+cudnn_enabled: true
+cudnn_benchmark: false
+cudnn_deterministic: false
+collect_stats: true
+write_collected_feats: false
+max_epoch: 130
+patience: null
+val_scheduler_criterion:
+- valid
+- loss
+early_stopping_criterion:
+- valid
+- loss
+- min
+best_model_criterion:
+-   - valid
+    - text2mel_loss
+    - min
+-   - train
+    - text2mel_loss
+    - min
+-   - train
+    - total_count
+    - max
+keep_nbest_models: -1
+nbest_averaging_interval: 0
+grad_clip: -1
+grad_clip_type: 2.0
+grad_noise: false
+accum_grad: 1
+no_forward_run: false
+resume: false
+train_dtype: float32
+use_amp: false
+log_interval: 50
+use_matplotlib: true
+use_tensorboard: true
+use_wandb: false
+wandb_project: null
+wandb_id: null
+wandb_entity: null
+wandb_name: null
+wandb_model_log_interval: -1
+detect_anomaly: false
+pretrain_path: null
+init_param: []
+ignore_init_mismatch: false
+freeze_param: []
+num_iters_per_epoch: 1000
+batch_size: 20
+valid_batch_size: null
+batch_bins: 6000000
+valid_batch_bins: null
+train_shape_file:
+- exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/train.2.scp
+valid_shape_file:
+- exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/valid.2.scp
+batch_type: numel
+valid_batch_type: null
+fold_length: []
+sort_in_batch: descending
+sort_batch: descending
+multiple_iterator: false
+chunk_length: 500
+chunk_shift_ratio: 0.5
+num_cache_chunks: 1024
+train_data_path_and_name_and_type:
+-   - dump/raw/jvs010_tr_no_dev/text
+    - text
+    - text
+-   - dump/raw/jvs010_tr_no_dev/wav.scp
+    - speech
+    - sound
+valid_data_path_and_name_and_type:
+-   - dump/raw/jvs010_dev/text
+    - text
+    - text
+-   - dump/raw/jvs010_dev/wav.scp
+    - speech
+    - sound
+allow_variable_data_keys: false
+max_cache_size: 0.0
+max_cache_fd: 32
+valid_max_cache_size: null
+optim: adamw
+optim_conf:
+    lr: 0.0002
+    betas:
+    - 0.8
+    - 0.99
+    eps: 1.0e-09
+    weight_decay: 0.0
+scheduler: exponentiallr
+scheduler_conf:
+    gamma: 0.999875
+optim2: adamw
+optim2_conf:
+    lr: 0.0002
+    betas:
+    - 0.8
+    - 0.99
+    eps: 1.0e-09
+    weight_decay: 0.0
+scheduler2: exponentiallr
+scheduler2_conf:
+    gamma: 0.999875
+generator_first: true
+token_list:
+- <blank>
+- <unk>
+- o
+- a
+- u
+- i
+- e
+- k
+- r
+- t
+- n
+- pau
+- N
+- s
+- sh
+- d
+- m
+- g
+- w
+- b
+- cl
+- I
+- j
+- ch
+- y
+- U
+- h
+- p
+- ts
+- f
+- z
+- ky
+- ny
+- gy
+- ry
+- hy
+- my
+- by
+- py
+- v
+- <sos/eos>
+odim: null
+model_conf: {}
+use_preprocessor: true
+token_type: phn
+bpemodel: null
+non_linguistic_symbols: null
+cleaner: jaconv
+g2p: pyopenjtalk
+feats_extract: fbank
+feats_extract_conf:
+    n_fft: 2048
+    hop_length: 300
+    win_length: 1200
+    fs: 24000
+    fmin: 80
+    fmax: 7600
+    n_mels: 80
+normalize: null
+normalize_conf: {}
+tts: jets
+tts_conf:
+    generator_type: jets_generator
+    generator_params:
+        adim: 256
+        aheads: 2
+        elayers: 4
+        eunits: 1024
+        dlayers: 4
+        dunits: 1024
+        positionwise_layer_type: conv1d
+        positionwise_conv_kernel_size: 3
+        duration_predictor_layers: 2
+        duration_predictor_chans: 256
+        duration_predictor_kernel_size: 3
+        use_masking: true
+        encoder_normalize_before: true
+        decoder_normalize_before: true
+        encoder_type: transformer
+        decoder_type: transformer
+        conformer_rel_pos_type: latest
+        conformer_pos_enc_layer_type: rel_pos
+        conformer_self_attn_layer_type: rel_selfattn
+        conformer_activation_type: swish
+        use_macaron_style_in_conformer: true
+        use_cnn_in_conformer: true
+        conformer_enc_kernel_size: 7
+        conformer_dec_kernel_size: 31
+        init_type: xavier_uniform
+        transformer_enc_dropout_rate: 0.2
+        transformer_enc_positional_dropout_rate: 0.2
+        transformer_enc_attn_dropout_rate: 0.2
+        transformer_dec_dropout_rate: 0.2
+        transformer_dec_positional_dropout_rate: 0.2
+        transformer_dec_attn_dropout_rate: 0.2
+        pitch_predictor_layers: 5
+        pitch_predictor_chans: 256
+        pitch_predictor_kernel_size: 5
+        pitch_predictor_dropout: 0.5
+        pitch_embed_kernel_size: 1
+        pitch_embed_dropout: 0.0
+        stop_gradient_from_pitch_predictor: true
+        energy_predictor_layers: 2
+        energy_predictor_chans: 256
+        energy_predictor_kernel_size: 3
+        energy_predictor_dropout: 0.5
+        energy_embed_kernel_size: 1
+        energy_embed_dropout: 0.0
+        stop_gradient_from_energy_predictor: false
+        generator_out_channels: 1
+        generator_channels: 512
+        generator_global_channels: -1
+        generator_kernel_size: 7
+        generator_upsample_scales:
+        - 8
+        - 8
+        - 2
+        - 2
+        generator_upsample_kernel_sizes:
+        - 16
+        - 16
+        - 4
+        - 4
+        generator_resblock_kernel_sizes:
+        - 3
+        - 7
+        - 11
+        generator_resblock_dilations:
+        -   - 1
+            - 3
+            - 5
+        -   - 1
+            - 3
+            - 5
+        -   - 1
+            - 3
+            - 5
+        generator_use_additional_convs: true
+        generator_bias: true
+        generator_nonlinear_activation: LeakyReLU
+        generator_nonlinear_activation_params:
+            negative_slope: 0.1
+        generator_use_weight_norm: true
+        segment_size: 64
+        idim: 41
+        odim: 80
+    discriminator_type: hifigan_multi_scale_multi_period_discriminator
+    discriminator_params:
+        scales: 1
+        scale_downsample_pooling: AvgPool1d
+        scale_downsample_pooling_params:
+            kernel_size: 4
+            stride: 2
+            padding: 2
+        scale_discriminator_params:
+            in_channels: 1
+            out_channels: 1
+            kernel_sizes:
+            - 15
+            - 41
+            - 5
+            - 3
+            channels: 128
+            max_downsample_channels: 1024
+            max_groups: 16
+            bias: true
+            downsample_scales:
+            - 2
+            - 2
+            - 4
+            - 4
+            - 1
+            nonlinear_activation: LeakyReLU
+            nonlinear_activation_params:
+                negative_slope: 0.1
+            use_weight_norm: true
+            use_spectral_norm: false
+        follow_official_norm: false
+        periods:
+        - 2
+        - 3
+        - 5
+        - 7
+        - 11
+        period_discriminator_params:
+            in_channels: 1
+            out_channels: 1
+            kernel_sizes:
+            - 5
+            - 3
+            channels: 32
+            downsample_scales:
+            - 3
+            - 3
+            - 3
+            - 3
+            - 1
+            max_downsample_channels: 1024
+            bias: true
+            nonlinear_activation: LeakyReLU
+            nonlinear_activation_params:
+                negative_slope: 0.1
+            use_weight_norm: true
+            use_spectral_norm: false
+    generator_adv_loss_params:
+        average_by_discriminators: false
+        loss_type: mse
+    discriminator_adv_loss_params:
+        average_by_discriminators: false
+        loss_type: mse
+    feat_match_loss_params:
+        average_by_discriminators: false
+        average_by_layers: false
+        include_final_outputs: true
+    mel_loss_params:
+        fs: 24000
+        n_fft: 1024
+        hop_length: 256
+        win_length: null
+        window: hann
+        n_mels: 80
+        fmin: 0
+        fmax: null
+        log_base: null
+    lambda_adv: 1.0
+    lambda_mel: 45.0
+    lambda_feat_match: 2.0
+    lambda_var: 1.0
+    lambda_align: 2.0
+    sampling_rate: 24000
+    cache_generator_outputs: true
+pitch_extract: dio
+pitch_extract_conf:
+    reduction_factor: 1
+    use_token_averaged_f0: false
+    fs: 24000
+    n_fft: 2048
+    hop_length: 300
+    f0max: 400
+    f0min: 80
+pitch_normalize: null
+pitch_normalize_conf: {}
+energy_extract: energy
+energy_extract_conf:
+    reduction_factor: 1
+    use_token_averaged_energy: false
+    fs: 24000
+    n_fft: 2048
+    hop_length: 300
+    win_length: 1200
+energy_normalize: null
+energy_normalize_conf: {}
+required:
+- output_dir
+- token_list
+version: '202204'
+distributed: false

exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/stats.2/train/batch_keys

@@ -0,0 +1,2 @@
+text
+speech

exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/stats.2/train/energy_lengths_stats.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:efe9bc4321cc000a1484565eb1b696bf9c1ed193a806ec8ebf606d4207b939d3
+size 778

exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/stats.2/train/energy_stats.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5fd01373077c58b6e8e5e20227f37180f96d8717bdf9c4e0bd17b93206fcd5a6
+size 770

exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/stats.2/train/feats_lengths_stats.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:efe9bc4321cc000a1484565eb1b696bf9c1ed193a806ec8ebf606d4207b939d3
+size 778

exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/stats.2/train/feats_stats.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0205589d93485e825bd25b9c6b5e1714ca6dae977b3f555591874bacaffed623
+size 1402

exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/stats.2/train/pitch_lengths_stats.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:efe9bc4321cc000a1484565eb1b696bf9c1ed193a806ec8ebf606d4207b939d3
+size 778

exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/stats.2/train/pitch_stats.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a418a92a86d6891b0a6ec4fbd84592a5df85011ea03995a231bb6e60f830f9ce
+size 770

exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/stats.2/train/speech_shape

@@ -0,0 +1,13 @@
+jvs010_VOICEACTRESS100_014 56160
+jvs010_VOICEACTRESS100_015 104640
+jvs010_VOICEACTRESS100_016 122880
+jvs010_VOICEACTRESS100_017 148080
+jvs010_VOICEACTRESS100_018 161041
+jvs010_VOICEACTRESS100_019 185040
+jvs010_VOICEACTRESS100_020 162000
+jvs010_VOICEACTRESS100_021 154320
+jvs010_VOICEACTRESS100_022 314880
+jvs010_VOICEACTRESS100_023 150480
+jvs010_VOICEACTRESS100_024 197040
+jvs010_VOICEACTRESS100_025 98880
+jvs010_VOICEACTRESS100_026 89040

exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/stats.2/train/stats_keys

@@ -0,0 +1,6 @@
+feats
+feats_lengths
+pitch
+pitch_lengths
+energy
+energy_lengths

exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/stats.2/train/text_shape

@@ -0,0 +1,13 @@
+jvs010_VOICEACTRESS100_014 33
+jvs010_VOICEACTRESS100_015 47
+jvs010_VOICEACTRESS100_016 62
+jvs010_VOICEACTRESS100_017 60
+jvs010_VOICEACTRESS100_018 78
+jvs010_VOICEACTRESS100_019 87
+jvs010_VOICEACTRESS100_020 80
+jvs010_VOICEACTRESS100_021 67
+jvs010_VOICEACTRESS100_022 124
+jvs010_VOICEACTRESS100_023 69
+jvs010_VOICEACTRESS100_024 86
+jvs010_VOICEACTRESS100_025 54
+jvs010_VOICEACTRESS100_026 40

exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/stats.2/valid/batch_keys

@@ -0,0 +1,2 @@
+text
+speech

exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/stats.2/valid/energy_lengths_stats.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:04927f2b066ae5e9114b3f175a0890ad05a42039ba563ddf956476dde6575140
+size 778

exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/stats.2/valid/energy_stats.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d7a5191cf7f4b801f806ca1d7f5b73c5d530a273d73d64158189a4bf9b8c4494
+size 770

exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/stats.2/valid/feats_lengths_stats.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:04927f2b066ae5e9114b3f175a0890ad05a42039ba563ddf956476dde6575140
+size 778

exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/stats.2/valid/feats_stats.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8766fdf1445b7fa6eaa6fffb246419dac1dac3315ee84b24a0546d58e5974aec
+size 1402

exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/stats.2/valid/pitch_lengths_stats.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:04927f2b066ae5e9114b3f175a0890ad05a42039ba563ddf956476dde6575140
+size 778

exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/stats.2/valid/pitch_stats.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2fed5ab4f8c3b3f81c11011b7f6d1eb8c0435c4c5a56361938accde0be1f3740
+size 770

exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/stats.2/valid/speech_shape

@@ -0,0 +1,2 @@
+jvs010_BASIC5000_0351 72720
+jvs010_BASIC5000_0882 91440

exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/stats.2/valid/stats_keys

@@ -0,0 +1,6 @@
+feats
+feats_lengths
+pitch
+pitch_lengths
+energy
+energy_lengths

exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/stats.2/valid/text_shape

@@ -0,0 +1,2 @@
+jvs010_BASIC5000_0351 36
+jvs010_BASIC5000_0882 49

exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/stats.3.log

@@ -0,0 +1,874 @@
+# python3 -m espnet2.bin.gan_tts_train --collect_stats true --write_collected_feats false --use_preprocessor true --token_type phn --token_list dump/token_list/phn_jaconv_pyopenjtalk/tokens.txt --non_linguistic_symbols none --cleaner jaconv --g2p pyopenjtalk --normalize none --pitch_normalize none --energy_normalize none --train_data_path_and_name_and_type dump/raw/jvs010_tr_no_dev/text,text,text --train_data_path_and_name_and_type dump/raw/jvs010_tr_no_dev/wav.scp,speech,sound --valid_data_path_and_name_and_type dump/raw/jvs010_dev/text,text,text --valid_data_path_and_name_and_type dump/raw/jvs010_dev/wav.scp,speech,sound --train_shape_file exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/train.3.scp --valid_shape_file exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/valid.3.scp --output_dir exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/stats.3 --config conf/tuning/train_jets.yaml --feats_extract fbank --feats_extract_conf n_fft=2048 --feats_extract_conf hop_length=300 --feats_extract_conf win_length=1200 --feats_extract_conf fs=24000 --feats_extract_conf fmin=80 --feats_extract_conf fmax=7600 --feats_extract_conf n_mels=80 --pitch_extract_conf fs=24000 --pitch_extract_conf n_fft=2048 --pitch_extract_conf hop_length=300 --pitch_extract_conf f0max=400 --pitch_extract_conf f0min=80 --energy_extract_conf fs=24000 --energy_extract_conf n_fft=2048 --energy_extract_conf hop_length=300 --energy_extract_conf win_length=1200 
+# Started at Tue Mar  4 21:23:26 JST 2025
+#
+/usr/lib/python3/dist-packages/requests/__init__.py:89: RequestsDependencyWarning: urllib3 (2.2.3) or chardet (3.0.4) doesn't match a supported version!
+  warnings.warn("urllib3 ({}) or chardet ({}) doesn't match a supported "
+/usr/bin/python3 /work/espnet/espnet2/bin/gan_tts_train.py --collect_stats true --write_collected_feats false --use_preprocessor true --token_type phn --token_list dump/token_list/phn_jaconv_pyopenjtalk/tokens.txt --non_linguistic_symbols none --cleaner jaconv --g2p pyopenjtalk --normalize none --pitch_normalize none --energy_normalize none --train_data_path_and_name_and_type dump/raw/jvs010_tr_no_dev/text,text,text --train_data_path_and_name_and_type dump/raw/jvs010_tr_no_dev/wav.scp,speech,sound --valid_data_path_and_name_and_type dump/raw/jvs010_dev/text,text,text --valid_data_path_and_name_and_type dump/raw/jvs010_dev/wav.scp,speech,sound --train_shape_file exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/train.3.scp --valid_shape_file exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/valid.3.scp --output_dir exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/stats.3 --config conf/tuning/train_jets.yaml --feats_extract fbank --feats_extract_conf n_fft=2048 --feats_extract_conf hop_length=300 --feats_extract_conf win_length=1200 --feats_extract_conf fs=24000 --feats_extract_conf fmin=80 --feats_extract_conf fmax=7600 --feats_extract_conf n_mels=80 --pitch_extract_conf fs=24000 --pitch_extract_conf n_fft=2048 --pitch_extract_conf hop_length=300 --pitch_extract_conf f0max=400 --pitch_extract_conf f0min=80 --energy_extract_conf fs=24000 --energy_extract_conf n_fft=2048 --energy_extract_conf hop_length=300 --energy_extract_conf win_length=1200
+[92b100c97f43] 2025-03-04 21:23:29,166 (gan_tts:304) INFO: Vocabulary size: 41
+[92b100c97f43] 2025-03-04 21:23:29,388 (encoder:172) INFO: encoder self-attention layer type = self-attention
+[92b100c97f43] 2025-03-04 21:23:29,512 (encoder:172) INFO: encoder self-attention layer type = self-attention
+[92b100c97f43] 2025-03-04 21:23:31,615 (abs_task:1157) INFO: pytorch.version=1.10.1+cu113, cuda.available=True, cudnn.version=8200, cudnn.benchmark=False, cudnn.deterministic=False
+[92b100c97f43] 2025-03-04 21:23:31,624 (abs_task:1158) INFO: Model structure:
+ESPnetGANTTSModel(
+  (feats_extract): LogMelFbank(
+    (stft): Stft(n_fft=2048, win_length=1200, hop_length=300, center=True, normalized=False, onesided=True)
+    (logmel): LogMel(sr=24000, n_fft=2048, n_mels=80, fmin=80, fmax=7600, htk=False)
+  )
+  (pitch_extract): Dio()
+  (energy_extract): Energy(
+    (stft): Stft(n_fft=2048, win_length=1200, hop_length=300, center=True, normalized=False, onesided=True)
+  )
+  (tts): JETS(
+    (generator): JETSGenerator(
+      (encoder): Encoder(
+        (embed): Sequential(
+          (0): Embedding(41, 256, padding_idx=0)
+          (1): ScaledPositionalEncoding(
+            (dropout): Dropout(p=0.2, inplace=False)
+          )
+        )
+        (encoders): MultiSequential(
+          (0): EncoderLayer(
+            (self_attn): MultiHeadedAttention(
+              (linear_q): Linear(in_features=256, out_features=256, bias=True)
+              (linear_k): Linear(in_features=256, out_features=256, bias=True)
+              (linear_v): Linear(in_features=256, out_features=256, bias=True)
+              (linear_out): Linear(in_features=256, out_features=256, bias=True)
+              (dropout): Dropout(p=0.2, inplace=False)
+            )
+            (feed_forward): MultiLayeredConv1d(
+              (w_1): Conv1d(256, 1024, kernel_size=(3,), stride=(1,), padding=(1,))
+              (w_2): Conv1d(1024, 256, kernel_size=(3,), stride=(1,), padding=(1,))
+              (dropout): Dropout(p=0.2, inplace=False)
+            )
+            (norm1): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (norm2): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (dropout): Dropout(p=0.2, inplace=False)
+          )
+          (1): EncoderLayer(
+            (self_attn): MultiHeadedAttention(
+              (linear_q): Linear(in_features=256, out_features=256, bias=True)
+              (linear_k): Linear(in_features=256, out_features=256, bias=True)
+              (linear_v): Linear(in_features=256, out_features=256, bias=True)
+              (linear_out): Linear(in_features=256, out_features=256, bias=True)
+              (dropout): Dropout(p=0.2, inplace=False)
+            )
+            (feed_forward): MultiLayeredConv1d(
+              (w_1): Conv1d(256, 1024, kernel_size=(3,), stride=(1,), padding=(1,))
+              (w_2): Conv1d(1024, 256, kernel_size=(3,), stride=(1,), padding=(1,))
+              (dropout): Dropout(p=0.2, inplace=False)
+            )
+            (norm1): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (norm2): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (dropout): Dropout(p=0.2, inplace=False)
+          )
+          (2): EncoderLayer(
+            (self_attn): MultiHeadedAttention(
+              (linear_q): Linear(in_features=256, out_features=256, bias=True)
+              (linear_k): Linear(in_features=256, out_features=256, bias=True)
+              (linear_v): Linear(in_features=256, out_features=256, bias=True)
+              (linear_out): Linear(in_features=256, out_features=256, bias=True)
+              (dropout): Dropout(p=0.2, inplace=False)
+            )
+            (feed_forward): MultiLayeredConv1d(
+              (w_1): Conv1d(256, 1024, kernel_size=(3,), stride=(1,), padding=(1,))
+              (w_2): Conv1d(1024, 256, kernel_size=(3,), stride=(1,), padding=(1,))
+              (dropout): Dropout(p=0.2, inplace=False)
+            )
+            (norm1): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (norm2): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (dropout): Dropout(p=0.2, inplace=False)
+          )
+          (3): EncoderLayer(
+            (self_attn): MultiHeadedAttention(
+              (linear_q): Linear(in_features=256, out_features=256, bias=True)
+              (linear_k): Linear(in_features=256, out_features=256, bias=True)
+              (linear_v): Linear(in_features=256, out_features=256, bias=True)
+              (linear_out): Linear(in_features=256, out_features=256, bias=True)
+              (dropout): Dropout(p=0.2, inplace=False)
+            )
+            (feed_forward): MultiLayeredConv1d(
+              (w_1): Conv1d(256, 1024, kernel_size=(3,), stride=(1,), padding=(1,))
+              (w_2): Conv1d(1024, 256, kernel_size=(3,), stride=(1,), padding=(1,))
+              (dropout): Dropout(p=0.2, inplace=False)
+            )
+            (norm1): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (norm2): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (dropout): Dropout(p=0.2, inplace=False)
+          )
+        )
+        (after_norm): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+      )
+      (duration_predictor): DurationPredictor(
+        (conv): ModuleList(
+          (0): Sequential(
+            (0): Conv1d(256, 256, kernel_size=(3,), stride=(1,), padding=(1,))
+            (1): ReLU()
+            (2): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (3): Dropout(p=0.1, inplace=False)
+          )
+          (1): Sequential(
+            (0): Conv1d(256, 256, kernel_size=(3,), stride=(1,), padding=(1,))
+            (1): ReLU()
+            (2): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (3): Dropout(p=0.1, inplace=False)
+          )
+        )
+        (linear): Linear(in_features=256, out_features=1, bias=True)
+      )
+      (pitch_predictor): VariancePredictor(
+        (conv): ModuleList(
+          (0): Sequential(
+            (0): Conv1d(256, 256, kernel_size=(5,), stride=(1,), padding=(2,))
+            (1): ReLU()
+            (2): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (3): Dropout(p=0.5, inplace=False)
+          )
+          (1): Sequential(
+            (0): Conv1d(256, 256, kernel_size=(5,), stride=(1,), padding=(2,))
+            (1): ReLU()
+            (2): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (3): Dropout(p=0.5, inplace=False)
+          )
+          (2): Sequential(
+            (0): Conv1d(256, 256, kernel_size=(5,), stride=(1,), padding=(2,))
+            (1): ReLU()
+            (2): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (3): Dropout(p=0.5, inplace=False)
+          )
+          (3): Sequential(
+            (0): Conv1d(256, 256, kernel_size=(5,), stride=(1,), padding=(2,))
+            (1): ReLU()
+            (2): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (3): Dropout(p=0.5, inplace=False)
+          )
+          (4): Sequential(
+            (0): Conv1d(256, 256, kernel_size=(5,), stride=(1,), padding=(2,))
+            (1): ReLU()
+            (2): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (3): Dropout(p=0.5, inplace=False)
+          )
+        )
+        (linear): Linear(in_features=256, out_features=1, bias=True)
+      )
+      (pitch_embed): Sequential(
+        (0): Conv1d(1, 256, kernel_size=(1,), stride=(1,))
+        (1): Dropout(p=0.0, inplace=False)
+      )
+      (energy_predictor): VariancePredictor(
+        (conv): ModuleList(
+          (0): Sequential(
+            (0): Conv1d(256, 256, kernel_size=(3,), stride=(1,), padding=(1,))
+            (1): ReLU()
+            (2): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (3): Dropout(p=0.5, inplace=False)
+          )
+          (1): Sequential(
+            (0): Conv1d(256, 256, kernel_size=(3,), stride=(1,), padding=(1,))
+            (1): ReLU()
+            (2): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (3): Dropout(p=0.5, inplace=False)
+          )
+        )
+        (linear): Linear(in_features=256, out_features=1, bias=True)
+      )
+      (energy_embed): Sequential(
+        (0): Conv1d(1, 256, kernel_size=(1,), stride=(1,))
+        (1): Dropout(p=0.0, inplace=False)
+      )
+      (alignment_module): AlignmentModule(
+        (t_conv1): Conv1d(256, 256, kernel_size=(3,), stride=(1,), padding=(1,))
+        (t_conv2): Conv1d(256, 256, kernel_size=(1,), stride=(1,))
+        (f_conv1): Conv1d(80, 256, kernel_size=(3,), stride=(1,), padding=(1,))
+        (f_conv2): Conv1d(256, 256, kernel_size=(3,), stride=(1,), padding=(1,))
+        (f_conv3): Conv1d(256, 256, kernel_size=(1,), stride=(1,))
+      )
+      (length_regulator): GaussianUpsampling()
+      (decoder): Encoder(
+        (embed): Sequential(
+          (0): ScaledPositionalEncoding(
+            (dropout): Dropout(p=0.2, inplace=False)
+          )
+        )
+        (encoders): MultiSequential(
+          (0): EncoderLayer(
+            (self_attn): MultiHeadedAttention(
+              (linear_q): Linear(in_features=256, out_features=256, bias=True)
+              (linear_k): Linear(in_features=256, out_features=256, bias=True)
+              (linear_v): Linear(in_features=256, out_features=256, bias=True)
+              (linear_out): Linear(in_features=256, out_features=256, bias=True)
+              (dropout): Dropout(p=0.2, inplace=False)
+            )
+            (feed_forward): MultiLayeredConv1d(
+              (w_1): Conv1d(256, 1024, kernel_size=(3,), stride=(1,), padding=(1,))
+              (w_2): Conv1d(1024, 256, kernel_size=(3,), stride=(1,), padding=(1,))
+              (dropout): Dropout(p=0.2, inplace=False)
+            )
+            (norm1): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (norm2): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (dropout): Dropout(p=0.2, inplace=False)
+          )
+          (1): EncoderLayer(
+            (self_attn): MultiHeadedAttention(
+              (linear_q): Linear(in_features=256, out_features=256, bias=True)
+              (linear_k): Linear(in_features=256, out_features=256, bias=True)
+              (linear_v): Linear(in_features=256, out_features=256, bias=True)
+              (linear_out): Linear(in_features=256, out_features=256, bias=True)
+              (dropout): Dropout(p=0.2, inplace=False)
+            )
+            (feed_forward): MultiLayeredConv1d(
+              (w_1): Conv1d(256, 1024, kernel_size=(3,), stride=(1,), padding=(1,))
+              (w_2): Conv1d(1024, 256, kernel_size=(3,), stride=(1,), padding=(1,))
+              (dropout): Dropout(p=0.2, inplace=False)
+            )
+            (norm1): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (norm2): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (dropout): Dropout(p=0.2, inplace=False)
+          )
+          (2): EncoderLayer(
+            (self_attn): MultiHeadedAttention(
+              (linear_q): Linear(in_features=256, out_features=256, bias=True)
+              (linear_k): Linear(in_features=256, out_features=256, bias=True)
+              (linear_v): Linear(in_features=256, out_features=256, bias=True)
+              (linear_out): Linear(in_features=256, out_features=256, bias=True)
+              (dropout): Dropout(p=0.2, inplace=False)
+            )
+            (feed_forward): MultiLayeredConv1d(
+              (w_1): Conv1d(256, 1024, kernel_size=(3,), stride=(1,), padding=(1,))
+              (w_2): Conv1d(1024, 256, kernel_size=(3,), stride=(1,), padding=(1,))
+              (dropout): Dropout(p=0.2, inplace=False)
+            )
+            (norm1): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (norm2): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (dropout): Dropout(p=0.2, inplace=False)
+          )
+          (3): EncoderLayer(
+            (self_attn): MultiHeadedAttention(
+              (linear_q): Linear(in_features=256, out_features=256, bias=True)
+              (linear_k): Linear(in_features=256, out_features=256, bias=True)
+              (linear_v): Linear(in_features=256, out_features=256, bias=True)
+              (linear_out): Linear(in_features=256, out_features=256, bias=True)
+              (dropout): Dropout(p=0.2, inplace=False)
+            )
+            (feed_forward): MultiLayeredConv1d(
+              (w_1): Conv1d(256, 1024, kernel_size=(3,), stride=(1,), padding=(1,))
+              (w_2): Conv1d(1024, 256, kernel_size=(3,), stride=(1,), padding=(1,))
+              (dropout): Dropout(p=0.2, inplace=False)
+            )
+            (norm1): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (norm2): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+            (dropout): Dropout(p=0.2, inplace=False)
+          )
+        )
+        (after_norm): LayerNorm((256,), eps=1e-12, elementwise_affine=True)
+      )
+      (generator): HiFiGANGenerator(
+        (input_conv): Conv1d(256, 512, kernel_size=(7,), stride=(1,), padding=(3,))
+        (upsamples): ModuleList(
+          (0): Sequential(
+            (0): LeakyReLU(negative_slope=0.1)
+            (1): ConvTranspose1d(512, 256, kernel_size=(16,), stride=(8,), padding=(4,))
+          )
+          (1): Sequential(
+            (0): LeakyReLU(negative_slope=0.1)
+            (1): ConvTranspose1d(256, 128, kernel_size=(16,), stride=(8,), padding=(4,))
+          )
+          (2): Sequential(
+            (0): LeakyReLU(negative_slope=0.1)
+            (1): ConvTranspose1d(128, 64, kernel_size=(4,), stride=(2,), padding=(1,))
+          )
+          (3): Sequential(
+            (0): LeakyReLU(negative_slope=0.1)
+            (1): ConvTranspose1d(64, 32, kernel_size=(4,), stride=(2,), padding=(1,))
+          )
+        )
+        (blocks): ModuleList(
+          (0): ResidualBlock(
+            (convs1): ModuleList(
+              (0): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(256, 256, kernel_size=(3,), stride=(1,), padding=(1,))
+              )
+              (1): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(256, 256, kernel_size=(3,), stride=(1,), padding=(3,), dilation=(3,))
+              )
+              (2): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(256, 256, kernel_size=(3,), stride=(1,), padding=(5,), dilation=(5,))
+              )
+            )
+            (convs2): ModuleList(
+              (0): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(256, 256, kernel_size=(3,), stride=(1,), padding=(1,))
+              )
+              (1): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(256, 256, kernel_size=(3,), stride=(1,), padding=(1,))
+              )
+              (2): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(256, 256, kernel_size=(3,), stride=(1,), padding=(1,))
+              )
+            )
+          )
+          (1): ResidualBlock(
+            (convs1): ModuleList(
+              (0): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(256, 256, kernel_size=(7,), stride=(1,), padding=(3,))
+              )
+              (1): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(256, 256, kernel_size=(7,), stride=(1,), padding=(9,), dilation=(3,))
+              )
+              (2): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(256, 256, kernel_size=(7,), stride=(1,), padding=(15,), dilation=(5,))
+              )
+            )
+            (convs2): ModuleList(
+              (0): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(256, 256, kernel_size=(7,), stride=(1,), padding=(3,))
+              )
+              (1): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(256, 256, kernel_size=(7,), stride=(1,), padding=(3,))
+              )
+              (2): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(256, 256, kernel_size=(7,), stride=(1,), padding=(3,))
+              )
+            )
+          )
+          (2): ResidualBlock(
+            (convs1): ModuleList(
+              (0): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(256, 256, kernel_size=(11,), stride=(1,), padding=(5,))
+              )
+              (1): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(256, 256, kernel_size=(11,), stride=(1,), padding=(15,), dilation=(3,))
+              )
+              (2): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(256, 256, kernel_size=(11,), stride=(1,), padding=(25,), dilation=(5,))
+              )
+            )
+            (convs2): ModuleList(
+              (0): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(256, 256, kernel_size=(11,), stride=(1,), padding=(5,))
+              )
+              (1): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(256, 256, kernel_size=(11,), stride=(1,), padding=(5,))
+              )
+              (2): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(256, 256, kernel_size=(11,), stride=(1,), padding=(5,))
+              )
+            )
+          )
+          (3): ResidualBlock(
+            (convs1): ModuleList(
+              (0): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(128, 128, kernel_size=(3,), stride=(1,), padding=(1,))
+              )
+              (1): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(128, 128, kernel_size=(3,), stride=(1,), padding=(3,), dilation=(3,))
+              )
+              (2): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(128, 128, kernel_size=(3,), stride=(1,), padding=(5,), dilation=(5,))
+              )
+            )
+            (convs2): ModuleList(
+              (0): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(128, 128, kernel_size=(3,), stride=(1,), padding=(1,))
+              )
+              (1): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(128, 128, kernel_size=(3,), stride=(1,), padding=(1,))
+              )
+              (2): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(128, 128, kernel_size=(3,), stride=(1,), padding=(1,))
+              )
+            )
+          )
+          (4): ResidualBlock(
+            (convs1): ModuleList(
+              (0): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(128, 128, kernel_size=(7,), stride=(1,), padding=(3,))
+              )
+              (1): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(128, 128, kernel_size=(7,), stride=(1,), padding=(9,), dilation=(3,))
+              )
+              (2): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(128, 128, kernel_size=(7,), stride=(1,), padding=(15,), dilation=(5,))
+              )
+            )
+            (convs2): ModuleList(
+              (0): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(128, 128, kernel_size=(7,), stride=(1,), padding=(3,))
+              )
+              (1): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(128, 128, kernel_size=(7,), stride=(1,), padding=(3,))
+              )
+              (2): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(128, 128, kernel_size=(7,), stride=(1,), padding=(3,))
+              )
+            )
+          )
+          (5): ResidualBlock(
+            (convs1): ModuleList(
+              (0): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(128, 128, kernel_size=(11,), stride=(1,), padding=(5,))
+              )
+              (1): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(128, 128, kernel_size=(11,), stride=(1,), padding=(15,), dilation=(3,))
+              )
+              (2): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(128, 128, kernel_size=(11,), stride=(1,), padding=(25,), dilation=(5,))
+              )
+            )
+            (convs2): ModuleList(
+              (0): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(128, 128, kernel_size=(11,), stride=(1,), padding=(5,))
+              )
+              (1): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(128, 128, kernel_size=(11,), stride=(1,), padding=(5,))
+              )
+              (2): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(128, 128, kernel_size=(11,), stride=(1,), padding=(5,))
+              )
+            )
+          )
+          (6): ResidualBlock(
+            (convs1): ModuleList(
+              (0): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(64, 64, kernel_size=(3,), stride=(1,), padding=(1,))
+              )
+              (1): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(64, 64, kernel_size=(3,), stride=(1,), padding=(3,), dilation=(3,))
+              )
+              (2): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(64, 64, kernel_size=(3,), stride=(1,), padding=(5,), dilation=(5,))
+              )
+            )
+            (convs2): ModuleList(
+              (0): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(64, 64, kernel_size=(3,), stride=(1,), padding=(1,))
+              )
+              (1): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(64, 64, kernel_size=(3,), stride=(1,), padding=(1,))
+              )
+              (2): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(64, 64, kernel_size=(3,), stride=(1,), padding=(1,))
+              )
+            )
+          )
+          (7): ResidualBlock(
+            (convs1): ModuleList(
+              (0): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(64, 64, kernel_size=(7,), stride=(1,), padding=(3,))
+              )
+              (1): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(64, 64, kernel_size=(7,), stride=(1,), padding=(9,), dilation=(3,))
+              )
+              (2): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(64, 64, kernel_size=(7,), stride=(1,), padding=(15,), dilation=(5,))
+              )
+            )
+            (convs2): ModuleList(
+              (0): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(64, 64, kernel_size=(7,), stride=(1,), padding=(3,))
+              )
+              (1): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(64, 64, kernel_size=(7,), stride=(1,), padding=(3,))
+              )
+              (2): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(64, 64, kernel_size=(7,), stride=(1,), padding=(3,))
+              )
+            )
+          )
+          (8): ResidualBlock(
+            (convs1): ModuleList(
+              (0): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(64, 64, kernel_size=(11,), stride=(1,), padding=(5,))
+              )
+              (1): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(64, 64, kernel_size=(11,), stride=(1,), padding=(15,), dilation=(3,))
+              )
+              (2): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(64, 64, kernel_size=(11,), stride=(1,), padding=(25,), dilation=(5,))
+              )
+            )
+            (convs2): ModuleList(
+              (0): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(64, 64, kernel_size=(11,), stride=(1,), padding=(5,))
+              )
+              (1): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(64, 64, kernel_size=(11,), stride=(1,), padding=(5,))
+              )
+              (2): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(64, 64, kernel_size=(11,), stride=(1,), padding=(5,))
+              )
+            )
+          )
+          (9): ResidualBlock(
+            (convs1): ModuleList(
+              (0): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(32, 32, kernel_size=(3,), stride=(1,), padding=(1,))
+              )
+              (1): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(32, 32, kernel_size=(3,), stride=(1,), padding=(3,), dilation=(3,))
+              )
+              (2): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(32, 32, kernel_size=(3,), stride=(1,), padding=(5,), dilation=(5,))
+              )
+            )
+            (convs2): ModuleList(
+              (0): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(32, 32, kernel_size=(3,), stride=(1,), padding=(1,))
+              )
+              (1): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(32, 32, kernel_size=(3,), stride=(1,), padding=(1,))
+              )
+              (2): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(32, 32, kernel_size=(3,), stride=(1,), padding=(1,))
+              )
+            )
+          )
+          (10): ResidualBlock(
+            (convs1): ModuleList(
+              (0): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(32, 32, kernel_size=(7,), stride=(1,), padding=(3,))
+              )
+              (1): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(32, 32, kernel_size=(7,), stride=(1,), padding=(9,), dilation=(3,))
+              )
+              (2): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(32, 32, kernel_size=(7,), stride=(1,), padding=(15,), dilation=(5,))
+              )
+            )
+            (convs2): ModuleList(
+              (0): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(32, 32, kernel_size=(7,), stride=(1,), padding=(3,))
+              )
+              (1): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(32, 32, kernel_size=(7,), stride=(1,), padding=(3,))
+              )
+              (2): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(32, 32, kernel_size=(7,), stride=(1,), padding=(3,))
+              )
+            )
+          )
+          (11): ResidualBlock(
+            (convs1): ModuleList(
+              (0): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(32, 32, kernel_size=(11,), stride=(1,), padding=(5,))
+              )
+              (1): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(32, 32, kernel_size=(11,), stride=(1,), padding=(15,), dilation=(3,))
+              )
+              (2): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(32, 32, kernel_size=(11,), stride=(1,), padding=(25,), dilation=(5,))
+              )
+            )
+            (convs2): ModuleList(
+              (0): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(32, 32, kernel_size=(11,), stride=(1,), padding=(5,))
+              )
+              (1): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(32, 32, kernel_size=(11,), stride=(1,), padding=(5,))
+              )
+              (2): Sequential(
+                (0): LeakyReLU(negative_slope=0.1)
+                (1): Conv1d(32, 32, kernel_size=(11,), stride=(1,), padding=(5,))
+              )
+            )
+          )
+        )
+        (output_conv): Sequential(
+          (0): LeakyReLU(negative_slope=0.01)
+          (1): Conv1d(32, 1, kernel_size=(7,), stride=(1,), padding=(3,))
+          (2): Tanh()
+        )
+      )
+    )
+    (discriminator): HiFiGANMultiScaleMultiPeriodDiscriminator(
+      (msd): HiFiGANMultiScaleDiscriminator(
+        (discriminators): ModuleList(
+          (0): HiFiGANScaleDiscriminator(
+            (layers): ModuleList(
+              (0): Sequential(
+                (0): Conv1d(1, 128, kernel_size=(15,), stride=(1,), padding=(7,))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (1): Sequential(
+                (0): Conv1d(128, 128, kernel_size=(41,), stride=(2,), padding=(20,), groups=4)
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (2): Sequential(
+                (0): Conv1d(128, 256, kernel_size=(41,), stride=(2,), padding=(20,), groups=16)
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (3): Sequential(
+                (0): Conv1d(256, 512, kernel_size=(41,), stride=(4,), padding=(20,), groups=16)
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (4): Sequential(
+                (0): Conv1d(512, 1024, kernel_size=(41,), stride=(4,), padding=(20,), groups=16)
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (5): Sequential(
+                (0): Conv1d(1024, 1024, kernel_size=(41,), stride=(1,), padding=(20,), groups=16)
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (6): Sequential(
+                (0): Conv1d(1024, 1024, kernel_size=(5,), stride=(1,), padding=(2,))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (7): Conv1d(1024, 1, kernel_size=(3,), stride=(1,), padding=(1,))
+            )
+          )
+        )
+      )
+      (mpd): HiFiGANMultiPeriodDiscriminator(
+        (discriminators): ModuleList(
+          (0): HiFiGANPeriodDiscriminator(
+            (convs): ModuleList(
+              (0): Sequential(
+                (0): Conv2d(1, 32, kernel_size=(5, 1), stride=(3, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (1): Sequential(
+                (0): Conv2d(32, 128, kernel_size=(5, 1), stride=(3, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (2): Sequential(
+                (0): Conv2d(128, 512, kernel_size=(5, 1), stride=(3, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (3): Sequential(
+                (0): Conv2d(512, 1024, kernel_size=(5, 1), stride=(3, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (4): Sequential(
+                (0): Conv2d(1024, 1024, kernel_size=(5, 1), stride=(1, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+            )
+            (output_conv): Conv2d(1024, 1, kernel_size=(2, 1), stride=(1, 1), padding=(1, 0))
+          )
+          (1): HiFiGANPeriodDiscriminator(
+            (convs): ModuleList(
+              (0): Sequential(
+                (0): Conv2d(1, 32, kernel_size=(5, 1), stride=(3, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (1): Sequential(
+                (0): Conv2d(32, 128, kernel_size=(5, 1), stride=(3, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (2): Sequential(
+                (0): Conv2d(128, 512, kernel_size=(5, 1), stride=(3, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (3): Sequential(
+                (0): Conv2d(512, 1024, kernel_size=(5, 1), stride=(3, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (4): Sequential(
+                (0): Conv2d(1024, 1024, kernel_size=(5, 1), stride=(1, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+            )
+            (output_conv): Conv2d(1024, 1, kernel_size=(2, 1), stride=(1, 1), padding=(1, 0))
+          )
+          (2): HiFiGANPeriodDiscriminator(
+            (convs): ModuleList(
+              (0): Sequential(
+                (0): Conv2d(1, 32, kernel_size=(5, 1), stride=(3, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (1): Sequential(
+                (0): Conv2d(32, 128, kernel_size=(5, 1), stride=(3, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (2): Sequential(
+                (0): Conv2d(128, 512, kernel_size=(5, 1), stride=(3, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (3): Sequential(
+                (0): Conv2d(512, 1024, kernel_size=(5, 1), stride=(3, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (4): Sequential(
+                (0): Conv2d(1024, 1024, kernel_size=(5, 1), stride=(1, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+            )
+            (output_conv): Conv2d(1024, 1, kernel_size=(2, 1), stride=(1, 1), padding=(1, 0))
+          )
+          (3): HiFiGANPeriodDiscriminator(
+            (convs): ModuleList(
+              (0): Sequential(
+                (0): Conv2d(1, 32, kernel_size=(5, 1), stride=(3, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (1): Sequential(
+                (0): Conv2d(32, 128, kernel_size=(5, 1), stride=(3, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (2): Sequential(
+                (0): Conv2d(128, 512, kernel_size=(5, 1), stride=(3, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (3): Sequential(
+                (0): Conv2d(512, 1024, kernel_size=(5, 1), stride=(3, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (4): Sequential(
+                (0): Conv2d(1024, 1024, kernel_size=(5, 1), stride=(1, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+            )
+            (output_conv): Conv2d(1024, 1, kernel_size=(2, 1), stride=(1, 1), padding=(1, 0))
+          )
+          (4): HiFiGANPeriodDiscriminator(
+            (convs): ModuleList(
+              (0): Sequential(
+                (0): Conv2d(1, 32, kernel_size=(5, 1), stride=(3, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (1): Sequential(
+                (0): Conv2d(32, 128, kernel_size=(5, 1), stride=(3, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (2): Sequential(
+                (0): Conv2d(128, 512, kernel_size=(5, 1), stride=(3, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (3): Sequential(
+                (0): Conv2d(512, 1024, kernel_size=(5, 1), stride=(3, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+              (4): Sequential(
+                (0): Conv2d(1024, 1024, kernel_size=(5, 1), stride=(1, 1), padding=(2, 0))
+                (1): LeakyReLU(negative_slope=0.1)
+              )
+            )
+            (output_conv): Conv2d(1024, 1, kernel_size=(2, 1), stride=(1, 1), padding=(1, 0))
+          )
+        )
+      )
+    )
+    (generator_adv_loss): GeneratorAdversarialLoss()
+    (discriminator_adv_loss): DiscriminatorAdversarialLoss()
+    (feat_match_loss): FeatureMatchLoss()
+    (mel_loss): MelSpectrogramLoss(
+      (wav_to_mel): LogMelFbank(
+        (stft): Stft(n_fft=1024, win_length=1024, hop_length=256, center=True, normalized=False, onesided=True)
+        (logmel): LogMel(sr=24000, n_fft=1024, n_mels=80, fmin=0, fmax=12000.0, htk=False)
+      )
+    )
+    (var_loss): VarianceLoss(
+      (mse_criterion): MSELoss()
+      (duration_criterion): DurationPredictorLoss(
+        (criterion): MSELoss()
+      )
+    )
+    (forwardsum_loss): ForwardSumLoss()
+  )
+)
+
+Model summary:
+    Class Name: ESPnetGANTTSModel
+    Total Number of model parameters: 83.28 M
+    Number of trainable parameters: 83.28 M (100.0%)
+    Size: 333.11 MB
+    Type: torch.float32
+[92b100c97f43] 2025-03-04 21:23:31,625 (abs_task:1161) INFO: Optimizer:
+AdamW (
+Parameter Group 0
+    amsgrad: False
+    betas: [0.8, 0.99]
+    eps: 1e-09
+    initial_lr: 0.0002
+    lr: 0.0002
+    weight_decay: 0.0
+)
+[92b100c97f43] 2025-03-04 21:23:31,625 (abs_task:1162) INFO: Scheduler: <torch.optim.lr_scheduler.ExponentialLR object at 0x7f725c8cd0a0>
+[92b100c97f43] 2025-03-04 21:23:31,625 (abs_task:1161) INFO: Optimizer2:
+AdamW (
+Parameter Group 0
+    amsgrad: False
+    betas: [0.8, 0.99]
+    eps: 1e-09
+    initial_lr: 0.0002
+    lr: 0.0002
+    weight_decay: 0.0
+)
+[92b100c97f43] 2025-03-04 21:23:31,625 (abs_task:1162) INFO: Scheduler2: <torch.optim.lr_scheduler.ExponentialLR object at 0x7f7278a78280>
+[92b100c97f43] 2025-03-04 21:23:31,625 (abs_task:1171) INFO: Saving the configuration in exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/stats.3/config.yaml
+[92b100c97f43] 2025-03-04 21:23:31,643 (abs_task:1182) INFO: Namespace(accum_grad=1, allow_variable_data_keys=False, batch_bins=6000000, batch_size=20, batch_type='numel', best_model_criterion=[['valid', 'text2mel_loss', 'min'], ['train', 'text2mel_loss', 'min'], ['train', 'total_count', 'max']], bpemodel=None, chunk_length=500, chunk_shift_ratio=0.5, cleaner='jaconv', collect_stats=True, config='conf/tuning/train_jets.yaml', cudnn_benchmark=False, cudnn_deterministic=False, cudnn_enabled=True, detect_anomaly=False, dist_backend='nccl', dist_init_method='env://', dist_launcher=None, dist_master_addr=None, dist_master_port=None, dist_rank=None, dist_world_size=None, distributed=False, dry_run=False, early_stopping_criterion=('valid', 'loss', 'min'), energy_extract='energy', energy_extract_conf={'reduction_factor': 1, 'use_token_averaged_energy': False, 'fs': 24000, 'n_fft': 2048, 'hop_length': 300, 'win_length': 1200}, energy_normalize=None, energy_normalize_conf={}, feats_extract='fbank', feats_extract_conf={'n_fft': 2048, 'hop_length': 300, 'win_length': 1200, 'fs': 24000, 'fmin': 80, 'fmax': 7600, 'n_mels': 80}, fold_length=[], freeze_param=[], g2p='pyopenjtalk', generator_first=True, grad_clip=-1, grad_clip_type=2.0, grad_noise=False, ignore_init_mismatch=False, init_param=[], iterator_type='sequence', keep_nbest_models=-1, local_rank=None, log_interval=50, log_level='INFO', max_cache_fd=32, max_cache_size=0.0, max_epoch=130, model_conf={}, multiple_iterator=False, multiprocessing_distributed=False, nbest_averaging_interval=0, ngpu=0, no_forward_run=False, non_linguistic_symbols=None, normalize=None, normalize_conf={}, num_att_plot=3, num_cache_chunks=1024, num_iters_per_epoch=1000, num_workers=16, odim=None, optim='adamw', optim2='adamw', optim2_conf={'lr': 0.0002, 'betas': [0.8, 0.99], 'eps': 1e-09, 'weight_decay': 0.0}, optim_conf={'lr': 0.0002, 'betas': [0.8, 0.99], 'eps': 1e-09, 'weight_decay': 0.0}, output_dir='exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/stats.3', patience=None, pitch_extract='dio', pitch_extract_conf={'reduction_factor': 1, 'use_token_averaged_f0': False, 'fs': 24000, 'n_fft': 2048, 'hop_length': 300, 'f0max': 400, 'f0min': 80}, pitch_normalize=None, pitch_normalize_conf={}, pretrain_path=None, print_config=False, required=['output_dir', 'token_list'], resume=False, scheduler='exponentiallr', scheduler2='exponentiallr', scheduler2_conf={'gamma': 0.999875}, scheduler_conf={'gamma': 0.999875}, seed=777, sharded_ddp=False, sort_batch='descending', sort_in_batch='descending', token_list=['<blank>', '<unk>', 'o', 'a', 'u', 'i', 'e', 'k', 'r', 't', 'n', 'pau', 'N', 's', 'sh', 'd', 'm', 'g', 'w', 'b', 'cl', 'I', 'j', 'ch', 'y', 'U', 'h', 'p', 'ts', 'f', 'z', 'ky', 'ny', 'gy', 'ry', 'hy', 'my', 'by', 'py', 'v', '<sos/eos>'], token_type='phn', train_data_path_and_name_and_type=[('dump/raw/jvs010_tr_no_dev/text', 'text', 'text'), ('dump/raw/jvs010_tr_no_dev/wav.scp', 'speech', 'sound')], train_dtype='float32', train_shape_file=['exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/train.3.scp'], tts='jets', tts_conf={'generator_type': 'jets_generator', 'generator_params': {'adim': 256, 'aheads': 2, 'elayers': 4, 'eunits': 1024, 'dlayers': 4, 'dunits': 1024, 'positionwise_layer_type': 'conv1d', 'positionwise_conv_kernel_size': 3, 'duration_predictor_layers': 2, 'duration_predictor_chans': 256, 'duration_predictor_kernel_size': 3, 'use_masking': True, 'encoder_normalize_before': True, 'decoder_normalize_before': True, 'encoder_type': 'transformer', 'decoder_type': 'transformer', 'conformer_rel_pos_type': 'latest', 'conformer_pos_enc_layer_type': 'rel_pos', 'conformer_self_attn_layer_type': 'rel_selfattn', 'conformer_activation_type': 'swish', 'use_macaron_style_in_conformer': True, 'use_cnn_in_conformer': True, 'conformer_enc_kernel_size': 7, 'conformer_dec_kernel_size': 31, 'init_type': 'xavier_uniform', 'transformer_enc_dropout_rate': 0.2, 'transformer_enc_positional_dropout_rate': 0.2, 'transformer_enc_attn_dropout_rate': 0.2, 'transformer_dec_dropout_rate': 0.2, 'transformer_dec_positional_dropout_rate': 0.2, 'transformer_dec_attn_dropout_rate': 0.2, 'pitch_predictor_layers': 5, 'pitch_predictor_chans': 256, 'pitch_predictor_kernel_size': 5, 'pitch_predictor_dropout': 0.5, 'pitch_embed_kernel_size': 1, 'pitch_embed_dropout': 0.0, 'stop_gradient_from_pitch_predictor': True, 'energy_predictor_layers': 2, 'energy_predictor_chans': 256, 'energy_predictor_kernel_size': 3, 'energy_predictor_dropout': 0.5, 'energy_embed_kernel_size': 1, 'energy_embed_dropout': 0.0, 'stop_gradient_from_energy_predictor': False, 'generator_out_channels': 1, 'generator_channels': 512, 'generator_global_channels': -1, 'generator_kernel_size': 7, 'generator_upsample_scales': [8, 8, 2, 2], 'generator_upsample_kernel_sizes': [16, 16, 4, 4], 'generator_resblock_kernel_sizes': [3, 7, 11], 'generator_resblock_dilations': [[1, 3, 5], [1, 3, 5], [1, 3, 5]], 'generator_use_additional_convs': True, 'generator_bias': True, 'generator_nonlinear_activation': 'LeakyReLU', 'generator_nonlinear_activation_params': {'negative_slope': 0.1}, 'generator_use_weight_norm': True, 'segment_size': 64, 'idim': 41, 'odim': 80}, 'discriminator_type': 'hifigan_multi_scale_multi_period_discriminator', 'discriminator_params': {'scales': 1, 'scale_downsample_pooling': 'AvgPool1d', 'scale_downsample_pooling_params': {'kernel_size': 4, 'stride': 2, 'padding': 2}, 'scale_discriminator_params': {'in_channels': 1, 'out_channels': 1, 'kernel_sizes': [15, 41, 5, 3], 'channels': 128, 'max_downsample_channels': 1024, 'max_groups': 16, 'bias': True, 'downsample_scales': [2, 2, 4, 4, 1], 'nonlinear_activation': 'LeakyReLU', 'nonlinear_activation_params': {'negative_slope': 0.1}, 'use_weight_norm': True, 'use_spectral_norm': False}, 'follow_official_norm': False, 'periods': [2, 3, 5, 7, 11], 'period_discriminator_params': {'in_channels': 1, 'out_channels': 1, 'kernel_sizes': [5, 3], 'channels': 32, 'downsample_scales': [3, 3, 3, 3, 1], 'max_downsample_channels': 1024, 'bias': True, 'nonlinear_activation': 'LeakyReLU', 'nonlinear_activation_params': {'negative_slope': 0.1}, 'use_weight_norm': True, 'use_spectral_norm': False}}, 'generator_adv_loss_params': {'average_by_discriminators': False, 'loss_type': 'mse'}, 'discriminator_adv_loss_params': {'average_by_discriminators': False, 'loss_type': 'mse'}, 'feat_match_loss_params': {'average_by_discriminators': False, 'average_by_layers': False, 'include_final_outputs': True}, 'mel_loss_params': {'fs': 24000, 'n_fft': 1024, 'hop_length': 256, 'win_length': None, 'window': 'hann', 'n_mels': 80, 'fmin': 0, 'fmax': None, 'log_base': None}, 'lambda_adv': 1.0, 'lambda_mel': 45.0, 'lambda_feat_match': 2.0, 'lambda_var': 1.0, 'lambda_align': 2.0, 'sampling_rate': 24000, 'cache_generator_outputs': True}, unused_parameters=True, use_amp=False, use_matplotlib=True, use_preprocessor=True, use_tensorboard=True, use_wandb=False, val_scheduler_criterion=('valid', 'loss'), valid_batch_bins=None, valid_batch_size=None, valid_batch_type=None, valid_data_path_and_name_and_type=[('dump/raw/jvs010_dev/text', 'text', 'text'), ('dump/raw/jvs010_dev/wav.scp', 'speech', 'sound')], valid_max_cache_size=None, valid_shape_file=['exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/valid.3.scp'], version='202204', wandb_entity=None, wandb_id=None, wandb_model_log_interval=-1, wandb_name=None, wandb_project=None, write_collected_feats=False)
+/work/espnet/espnet2/layers/stft.py:166: UserWarning: __floordiv__ is deprecated, and its behavior will change in a future version of pytorch. It currently rounds toward 0 (like the 'trunc' function NOT 'floor'). This results in incorrect rounding for negative values. To keep the current behavior, use torch.div(a, b, rounding_mode='trunc'), or for actual floor division, use torch.div(a, b, rounding_mode='floor').
+  olens = (ilens - self.n_fft) // self.hop_length + 1
+# Accounting: time=11 threads=1
+# Ended (code 0) at Tue Mar  4 21:23:37 JST 2025, elapsed time 11 seconds

exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/stats.3/config.yaml

@@ -0,0 +1,386 @@
+config: conf/tuning/train_jets.yaml
+print_config: false
+log_level: INFO
+dry_run: false
+iterator_type: sequence
+output_dir: exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/stats.3
+ngpu: 0
+seed: 777
+num_workers: 16
+num_att_plot: 3
+dist_backend: nccl
+dist_init_method: env://
+dist_world_size: null
+dist_rank: null
+local_rank: null
+dist_master_addr: null
+dist_master_port: null
+dist_launcher: null
+multiprocessing_distributed: false
+unused_parameters: true
+sharded_ddp: false
+cudnn_enabled: true
+cudnn_benchmark: false
+cudnn_deterministic: false
+collect_stats: true
+write_collected_feats: false
+max_epoch: 130
+patience: null
+val_scheduler_criterion:
+- valid
+- loss
+early_stopping_criterion:
+- valid
+- loss
+- min
+best_model_criterion:
+-   - valid
+    - text2mel_loss
+    - min
+-   - train
+    - text2mel_loss
+    - min
+-   - train
+    - total_count
+    - max
+keep_nbest_models: -1
+nbest_averaging_interval: 0
+grad_clip: -1
+grad_clip_type: 2.0
+grad_noise: false
+accum_grad: 1
+no_forward_run: false
+resume: false
+train_dtype: float32
+use_amp: false
+log_interval: 50
+use_matplotlib: true
+use_tensorboard: true
+use_wandb: false
+wandb_project: null
+wandb_id: null
+wandb_entity: null
+wandb_name: null
+wandb_model_log_interval: -1
+detect_anomaly: false
+pretrain_path: null
+init_param: []
+ignore_init_mismatch: false
+freeze_param: []
+num_iters_per_epoch: 1000
+batch_size: 20
+valid_batch_size: null
+batch_bins: 6000000
+valid_batch_bins: null
+train_shape_file:
+- exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/train.3.scp
+valid_shape_file:
+- exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/valid.3.scp
+batch_type: numel
+valid_batch_type: null
+fold_length: []
+sort_in_batch: descending
+sort_batch: descending
+multiple_iterator: false
+chunk_length: 500
+chunk_shift_ratio: 0.5
+num_cache_chunks: 1024
+train_data_path_and_name_and_type:
+-   - dump/raw/jvs010_tr_no_dev/text
+    - text
+    - text
+-   - dump/raw/jvs010_tr_no_dev/wav.scp
+    - speech
+    - sound
+valid_data_path_and_name_and_type:
+-   - dump/raw/jvs010_dev/text
+    - text
+    - text
+-   - dump/raw/jvs010_dev/wav.scp
+    - speech
+    - sound
+allow_variable_data_keys: false
+max_cache_size: 0.0
+max_cache_fd: 32
+valid_max_cache_size: null
+optim: adamw
+optim_conf:
+    lr: 0.0002
+    betas:
+    - 0.8
+    - 0.99
+    eps: 1.0e-09
+    weight_decay: 0.0
+scheduler: exponentiallr
+scheduler_conf:
+    gamma: 0.999875
+optim2: adamw
+optim2_conf:
+    lr: 0.0002
+    betas:
+    - 0.8
+    - 0.99
+    eps: 1.0e-09
+    weight_decay: 0.0
+scheduler2: exponentiallr
+scheduler2_conf:
+    gamma: 0.999875
+generator_first: true
+token_list:
+- <blank>
+- <unk>
+- o
+- a
+- u
+- i
+- e
+- k
+- r
+- t
+- n
+- pau
+- N
+- s
+- sh
+- d
+- m
+- g
+- w
+- b
+- cl
+- I
+- j
+- ch
+- y
+- U
+- h
+- p
+- ts
+- f
+- z
+- ky
+- ny
+- gy
+- ry
+- hy
+- my
+- by
+- py
+- v
+- <sos/eos>
+odim: null
+model_conf: {}
+use_preprocessor: true
+token_type: phn
+bpemodel: null
+non_linguistic_symbols: null
+cleaner: jaconv
+g2p: pyopenjtalk
+feats_extract: fbank
+feats_extract_conf:
+    n_fft: 2048
+    hop_length: 300
+    win_length: 1200
+    fs: 24000
+    fmin: 80
+    fmax: 7600
+    n_mels: 80
+normalize: null
+normalize_conf: {}
+tts: jets
+tts_conf:
+    generator_type: jets_generator
+    generator_params:
+        adim: 256
+        aheads: 2
+        elayers: 4
+        eunits: 1024
+        dlayers: 4
+        dunits: 1024
+        positionwise_layer_type: conv1d
+        positionwise_conv_kernel_size: 3
+        duration_predictor_layers: 2
+        duration_predictor_chans: 256
+        duration_predictor_kernel_size: 3
+        use_masking: true
+        encoder_normalize_before: true
+        decoder_normalize_before: true
+        encoder_type: transformer
+        decoder_type: transformer
+        conformer_rel_pos_type: latest
+        conformer_pos_enc_layer_type: rel_pos
+        conformer_self_attn_layer_type: rel_selfattn
+        conformer_activation_type: swish
+        use_macaron_style_in_conformer: true
+        use_cnn_in_conformer: true
+        conformer_enc_kernel_size: 7
+        conformer_dec_kernel_size: 31
+        init_type: xavier_uniform
+        transformer_enc_dropout_rate: 0.2
+        transformer_enc_positional_dropout_rate: 0.2
+        transformer_enc_attn_dropout_rate: 0.2
+        transformer_dec_dropout_rate: 0.2
+        transformer_dec_positional_dropout_rate: 0.2
+        transformer_dec_attn_dropout_rate: 0.2
+        pitch_predictor_layers: 5
+        pitch_predictor_chans: 256
+        pitch_predictor_kernel_size: 5
+        pitch_predictor_dropout: 0.5
+        pitch_embed_kernel_size: 1
+        pitch_embed_dropout: 0.0
+        stop_gradient_from_pitch_predictor: true
+        energy_predictor_layers: 2
+        energy_predictor_chans: 256
+        energy_predictor_kernel_size: 3
+        energy_predictor_dropout: 0.5
+        energy_embed_kernel_size: 1
+        energy_embed_dropout: 0.0
+        stop_gradient_from_energy_predictor: false
+        generator_out_channels: 1
+        generator_channels: 512
+        generator_global_channels: -1
+        generator_kernel_size: 7
+        generator_upsample_scales:
+        - 8
+        - 8
+        - 2
+        - 2
+        generator_upsample_kernel_sizes:
+        - 16
+        - 16
+        - 4
+        - 4
+        generator_resblock_kernel_sizes:
+        - 3
+        - 7
+        - 11
+        generator_resblock_dilations:
+        -   - 1
+            - 3
+            - 5
+        -   - 1
+            - 3
+            - 5
+        -   - 1
+            - 3
+            - 5
+        generator_use_additional_convs: true
+        generator_bias: true
+        generator_nonlinear_activation: LeakyReLU
+        generator_nonlinear_activation_params:
+            negative_slope: 0.1
+        generator_use_weight_norm: true
+        segment_size: 64
+        idim: 41
+        odim: 80
+    discriminator_type: hifigan_multi_scale_multi_period_discriminator
+    discriminator_params:
+        scales: 1
+        scale_downsample_pooling: AvgPool1d
+        scale_downsample_pooling_params:
+            kernel_size: 4
+            stride: 2
+            padding: 2
+        scale_discriminator_params:
+            in_channels: 1
+            out_channels: 1
+            kernel_sizes:
+            - 15
+            - 41
+            - 5
+            - 3
+            channels: 128
+            max_downsample_channels: 1024
+            max_groups: 16
+            bias: true
+            downsample_scales:
+            - 2
+            - 2
+            - 4
+            - 4
+            - 1
+            nonlinear_activation: LeakyReLU
+            nonlinear_activation_params:
+                negative_slope: 0.1
+            use_weight_norm: true
+            use_spectral_norm: false
+        follow_official_norm: false
+        periods:
+        - 2
+        - 3
+        - 5
+        - 7
+        - 11
+        period_discriminator_params:
+            in_channels: 1
+            out_channels: 1
+            kernel_sizes:
+            - 5
+            - 3
+            channels: 32
+            downsample_scales:
+            - 3
+            - 3
+            - 3
+            - 3
+            - 1
+            max_downsample_channels: 1024
+            bias: true
+            nonlinear_activation: LeakyReLU
+            nonlinear_activation_params:
+                negative_slope: 0.1
+            use_weight_norm: true
+            use_spectral_norm: false
+    generator_adv_loss_params:
+        average_by_discriminators: false
+        loss_type: mse
+    discriminator_adv_loss_params:
+        average_by_discriminators: false
+        loss_type: mse
+    feat_match_loss_params:
+        average_by_discriminators: false
+        average_by_layers: false
+        include_final_outputs: true
+    mel_loss_params:
+        fs: 24000
+        n_fft: 1024
+        hop_length: 256
+        win_length: null
+        window: hann
+        n_mels: 80
+        fmin: 0
+        fmax: null
+        log_base: null
+    lambda_adv: 1.0
+    lambda_mel: 45.0
+    lambda_feat_match: 2.0
+    lambda_var: 1.0
+    lambda_align: 2.0
+    sampling_rate: 24000
+    cache_generator_outputs: true
+pitch_extract: dio
+pitch_extract_conf:
+    reduction_factor: 1
+    use_token_averaged_f0: false
+    fs: 24000
+    n_fft: 2048
+    hop_length: 300
+    f0max: 400
+    f0min: 80
+pitch_normalize: null
+pitch_normalize_conf: {}
+energy_extract: energy
+energy_extract_conf:
+    reduction_factor: 1
+    use_token_averaged_energy: false
+    fs: 24000
+    n_fft: 2048
+    hop_length: 300
+    win_length: 1200
+energy_normalize: null
+energy_normalize_conf: {}
+required:
+- output_dir
+- token_list
+version: '202204'
+distributed: false

exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/stats.3/train/batch_keys

@@ -0,0 +1,2 @@
+text
+speech

exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/stats.3/train/energy_lengths_stats.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ae6a03be1e1fb8a7ead4429f0b519db24a8ba45dab0a18a87835ea8affbc2e85
+size 778

exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/stats.3/train/energy_stats.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9ff60189b8d9d8b107ce989ba373218b4d36d74fd9e5e3b44cd70e59d9d40c97
+size 770

exp/tts_stats_raw_phn_jaconv_pyopenjtalk/logdir/stats.3/train/feats_lengths_stats.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ae6a03be1e1fb8a7ead4429f0b519db24a8ba45dab0a18a87835ea8affbc2e85
+size 778