add pretrained model

README.md

@@ -1,3 +1,18 @@
 ---
+tags:
+- espnet
+- audio
+- singing-voice-synthesis
+language: zh
+datasets:
+- opencpop
 license: cc-by-4.0
 ---
+
+## ESPnet2 SVS model
+
+### `espnet/opencpop2_svs2_toksing_pretrain`
+
+This model was trained by TangRain using opencpop recipe in [espnet](https://github.com/espnet/espnet/).
+
+Train the model for 300 epochs and choose the one **with the best performance based on validation loss**.

exp/svs_stats_raw_phn_none_zh/train/feats_stats.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:45c819a05078a3ba122477fdbd6d60394a519fb9fd1116e75dbe19531b9251ab
+size 1402

exp/svs_stats_raw_phn_none_zh/train/pitch_stats.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c62642bb93923108eb02f125bd653d06816bdfd4f6d66091468552360e6a01c6
+size 770

exp/svs_train_discrete_acoustic_raw_phn_none_zh/49epoch.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:51065dfe6873cb5078ac43aabc8160789057645454af1bb463e3a5af05722df4
+size 281016849

exp/svs_train_discrete_acoustic_raw_phn_none_zh/config.yaml

@@ -0,0 +1,304 @@
+config: conf/tuning/train_discrete_acoustic.yaml
+print_config: false
+log_level: INFO
+drop_last_iter: false
+dry_run: false
+iterator_type: sequence
+valid_iterator_type: null
+output_dir: exp/svs_train_discrete_acoustic_raw_phn_none_zh
+ngpu: 1
+seed: 0
+num_workers: 10
+num_att_plot: 3
+dist_backend: nccl
+dist_init_method: env://
+dist_world_size: null
+dist_rank: null
+local_rank: 0
+dist_master_addr: null
+dist_master_port: null
+dist_launcher: null
+multiprocessing_distributed: false
+unused_parameters: false
+sharded_ddp: false
+cudnn_enabled: true
+cudnn_benchmark: false
+cudnn_deterministic: true
+collect_stats: false
+write_collected_feats: false
+max_epoch: 300
+patience: null
+val_scheduler_criterion:
+- valid
+- loss
+early_stopping_criterion:
+- valid
+- loss
+- min
+best_model_criterion:
+-   - valid
+    - loss
+    - min
+-   - train
+    - loss
+    - min
+keep_nbest_models: 5
+nbest_averaging_interval: 0
+grad_clip: 1.0
+grad_clip_type: 2.0
+grad_noise: false
+accum_grad: 1
+no_forward_run: false
+resume: true
+train_dtype: float32
+use_amp: false
+log_interval: null
+use_matplotlib: true
+use_tensorboard: true
+create_graph_in_tensorboard: false
+use_wandb: false
+wandb_project: null
+wandb_id: null
+wandb_entity: null
+wandb_name: null
+wandb_model_log_interval: -1
+detect_anomaly: false
+use_lora: false
+save_lora_only: true
+lora_conf: {}
+pretrain_path: null
+init_param: []
+ignore_init_mismatch: false
+freeze_param: []
+num_iters_per_epoch: null
+batch_size: 32
+valid_batch_size: null
+batch_bins: 1000000
+valid_batch_bins: null
+train_shape_file:
+- exp/svs_stats_raw_phn_none_zh/train/text_shape.phn
+- exp/svs_stats_raw_phn_none_zh/train/singing_shape
+valid_shape_file:
+- exp/svs_stats_raw_phn_none_zh/valid/text_shape.phn
+- exp/svs_stats_raw_phn_none_zh/valid/singing_shape
+batch_type: sorted
+valid_batch_type: null
+fold_length:
+- 150
+- 256000
+sort_in_batch: descending
+shuffle_within_batch: false
+sort_batch: descending
+multiple_iterator: false
+chunk_length: 500
+chunk_shift_ratio: 0.5
+num_cache_chunks: 1024
+chunk_excluded_key_prefixes: []
+train_data_path_and_name_and_type:
+-   - dump/raw/tr_no_dev/text
+    - text
+    - text
+-   - dump/raw/tr_no_dev/wav.scp
+    - singing
+    - sound
+-   - dump/raw/tr_no_dev/label
+    - label
+    - duration
+-   - dump/raw/tr_no_dev/score.scp
+    - score
+    - score
+-   - dump/raw/tr_no_dev/token_multi_wavlm_large_6+wavlm_large_23+hubert_large_6
+    - discrete_token
+    - text_int
+valid_data_path_and_name_and_type:
+-   - dump/raw/dev/text
+    - text
+    - text
+-   - dump/raw/dev/wav.scp
+    - singing
+    - sound
+-   - dump/raw/dev/label
+    - label
+    - duration
+-   - dump/raw/dev/score.scp
+    - score
+    - score
+-   - dump/raw/dev/token_multi_wavlm_large_6+wavlm_large_23+hubert_large_6
+    - discrete_token
+    - text_int
+allow_variable_data_keys: false
+max_cache_size: 0.0
+max_cache_fd: 32
+allow_multi_rates: false
+valid_max_cache_size: null
+exclude_weight_decay: false
+exclude_weight_decay_conf: {}
+optim: adam
+optim_conf:
+    lr: 0.001
+    eps: 1.0e-06
+    weight_decay: 0.0
+scheduler: null
+scheduler_conf: {}
+token_list:
+- <blank>
+- <unk>
+- SP
+- i
+- AP
+- e
+- y
+- d
+- w
+- sh
+- ai
+- n
+- x
+- j
+- ian
+- u
+- l
+- h
+- b
+- o
+- zh
+- an
+- ou
+- m
+- q
+- z
+- en
+- g
+- ing
+- ei
+- ao
+- ang
+- uo
+- eng
+- t
+- a
+- ong
+- ui
+- k
+- f
+- r
+- iang
+- ch
+- v
+- in
+- iao
+- ie
+- iu
+- c
+- s
+- van
+- p
+- ve
+- uan
+- uang
+- ia
+- ua
+- uai
+- un
+- er
+- vn
+- iong
+- <sos/eos>
+odim: null
+model_conf: {}
+use_preprocessor: true
+token_type: phn
+bpemodel: null
+non_linguistic_symbols: null
+cleaner: null
+g2p: null
+fs: 16000
+discrete_token_layers: 3
+nclusters: 128
+score_feats_extract: syllable_score_feats
+score_feats_extract_conf:
+    fs: 16000
+    n_fft: 2048
+    win_length: 1200
+    hop_length: 320
+feats_extract: fbank
+feats_extract_conf:
+    n_fft: 2048
+    hop_length: 320
+    win_length: 1200
+    fs: 16000
+    fmin: 80
+    fmax: 7600
+    n_mels: 80
+normalize: global_mvn
+normalize_conf:
+    stats_file: exp/svs_stats_raw_phn_none_zh/train/feats_stats.npz
+svs: discrete_acoustic
+svs_conf:
+    midi_dim: 129
+    duration_dim: 512
+    adim: 384
+    aheads: 4
+    elayers: 6
+    eunits: 1536
+    dlayers: 6
+    dunits: 1536
+    postnet_layers: 0
+    postnet_chans: 512
+    postnet_filts: 5
+    postnet_dropout_rate: 0.5
+    use_batch_norm: true
+    reduction_factor: 1
+    global_channels: -1
+    text_encoder_attention_heads: 2
+    text_encoder_ffn_expand: 4
+    text_encoder_blocks: 6
+    text_encoder_positionwise_layer_type: conv1d
+    text_encoder_positionwise_conv_kernel_size: 3
+    text_encoder_positional_encoding_layer_type: rel_pos
+    text_encoder_self_attention_layer_type: rel_selfattn
+    text_encoder_activation_type: swish
+    text_encoder_normalize_before: true
+    text_encoder_dropout_rate: 0.1
+    text_encoder_positional_dropout_rate: 0.0
+    text_encoder_attention_dropout_rate: 0.1
+    use_macaron_style_in_text_encoder: true
+    use_conformer_conv_in_text_encoder: false
+    text_encoder_conformer_kernel_size: -1
+    init_type: pytorch
+    use_masking: true
+    loss_function: FastSpeech1
+    loss_type: L1
+    lambda_out: 1
+    lambda_dur: 1
+    lambda_pitch: 1
+    lambda_vuv: 0.01
+    use_discrete_token: true
+    predict_pitch: true
+    codec_codebook: 0
+pitch_extract: dio
+pitch_extract_conf:
+    use_token_averaged_f0: false
+    use_log_f0: true
+    fs: 16000
+    n_fft: 2048
+    hop_length: 320
+    f0max: 800
+    f0min: 80
+    reduction_factor: 1
+pitch_normalize: null
+pitch_normalize_conf:
+    stats_file: exp/svs_stats_raw_phn_none_zh/train/pitch_stats.npz
+ying_extract: null
+ying_extract_conf: {}
+energy_extract: null
+energy_extract_conf: {}
+energy_normalize: null
+energy_normalize_conf: {}
+model_type: discrete_svs
+model_type_conf: {}
+required:
+- output_dir
+- token_list
+version: '202310'
+distributed: false