Upload tiny_tts/utils/config.py with huggingface_hub

tiny_tts/utils/config.py

@@ -5,18 +5,19 @@ HOP_LENGTH = 512
 SEGMENT_FRAMES = 32
 ADD_BLANK = True
 SPEC_CHANNELS = FILTER_LENGTH // 2 + 1  # 1025
+N_MEL_CHANNELS = 128                      # updated in new checkpoint
 
 # Speakers
 N_SPEAKERS = 1
-SPK2ID = {"female": 0}
+SPK2ID = {"MALE": 0}
 
-# Model
+# Model — matches config.json for G_150000.pth (lighter version)
 MODEL_PARAMS = dict(
     use_spk_conditioned_encoder=True,
     use_noise_scaled_mas=True,
-    inter_channels=80,
-    hidden_channels=80,
-    filter_channels=320,
+    inter_channels=32,
+    hidden_channels=32,
+    filter_channels=128,
     n_heads=2,
     n_layers=3,
     n_layers_trans_flow=3,
@@ -26,16 +27,16 @@ MODEL_PARAMS = dict(
     resblock_kernel_sizes=[3, 7, 11],
     resblock_dilation_sizes=[[1, 3, 5], [1, 3, 5], [1, 3, 5]],
     upsample_rates=[8, 8, 2, 2, 2],
-    upsample_initial_channel=256,
+    upsample_initial_channel=64,
     upsample_kernel_sizes=[16, 16, 8, 2, 2],
     n_layers_q=3,
     use_spectral_norm=False,
-    gin_channels=80,
+    gin_channels=128,
     use_sdp=True,
     mas_noise_scale_initial=0.01,
     noise_scale_delta=2e-06,
 )
 
 # Language / Tone
-NUM_LANGUAGES = 10
-NUM_TONES = 16
+NUM_LANGUAGES = 1
+NUM_TONES = 6