azalea/VITS-fast-fine-tuning
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

upload files

Browse Source
This commit is contained in:
Plachta
2023-02-13 16:27:18 +08:00
parent 8fe8c019e8
commit ea90485218
15 changed files with 324 additions and 750 deletions
Download Patch File Download Diff File Expand all files Collapse all files
configs/chinese_base.json
-55
View File
@@ -1,55 +0,0 @@
{
"train": {
"log_interval": 200,
"eval_interval": 1000,
"seed": 1234,
"epochs": 10000,
"learning_rate": 2e-4,
"betas": [0.8, 0.99],
"eps": 1e-9,
"batch_size": 32,
"fp16_run": true,
"lr_decay": 0.999875,
"segment_size": 8192,
"init_lr_ratio": 1,
"warmup_epochs": 0,
"c_mel": 45,
"c_kl": 1.0
},
"data": {
"training_files":"filelists/juzi_train_filelist.txt.cleaned",
"validation_files":"filelists/juzi_val_filelist.txt.cleaned",
"text_cleaners":["chinese_cleaners"],
"max_wav_value": 32768.0,
"sampling_rate": 22050,
"filter_length": 1024,
"hop_length": 256,
"win_length": 1024,
"n_mel_channels": 80,
"mel_fmin": 0.0,
"mel_fmax": null,
"add_blank": true,
"n_speakers": 8,
"cleaned_text": true
},
"model": {
"inter_channels": 192,
"hidden_channels": 192,
"filter_channels": 768,
"n_heads": 2,
"n_layers": 6,
"kernel_size": 3,
"p_dropout": 0.1,
"resblock": "1",
"resblock_kernel_sizes": [3,7,11],
"resblock_dilation_sizes": [[1,3,5], [1,3,5], [1,3,5]],
"upsample_rates": [8,8,2,2],
"upsample_initial_channel": 512,
"upsample_kernel_sizes": [16,16,4,4],
"n_layers_q": 3,
"use_spectral_norm": false,
"gin_channels": 256
},
"speakers": ["\u5c0f\u8338", "\u5510\u4e50\u541f", "\u5c0f\u6bb7", "\u82b1\u73b2", "\u8bb8\u8001\u5e08", "\u90b1\u7433", "\u4e03\u4e00", "\u516b\u56db"],
"symbols": ["_", "\uff0c", "\u3002", "\uff01", "\uff1f", "\u2014", "\u2026", "\u3105", "\u3106", "\u3107", "\u3108", "\u3109", "\u310a", "\u310b", "\u310c", "\u310d", "\u310e", "\u310f", "\u3110", "\u3111", "\u3112", "\u3113", "\u3114", "\u3115", "\u3116", "\u3117", "\u3118", "\u3119", "\u311a", "\u311b", "\u311c", "\u311d", "\u311e", "\u311f", "\u3120", "\u3121", "\u3122", "\u3123", "\u3124", "\u3125", "\u3126", "\u3127", "\u3128", "\u3129", "\u02c9", "\u02ca", "\u02c7", "\u02cb", "\u02d9", " "]
}
configs/cjke_base.json
-54
View File
@@ -1,54 +0,0 @@
{
"train": {
"log_interval": 200,
"eval_interval": 1000,
"seed": 1234,
"epochs": 10000,
"learning_rate": 2e-4,
"betas": [0.8, 0.99],
"eps": 1e-9,
"batch_size": 32,
"fp16_run": true,
"lr_decay": 0.999875,
"segment_size": 8192,
"init_lr_ratio": 1,
"warmup_epochs": 0,
"c_mel": 45,
"c_kl": 1.0
},
"data": {
"training_files":"filelists/cjke_train_filelist.txt.cleaned",
"validation_files":"filelists/cjke_val_filelist.txt.cleaned",
"text_cleaners":["cjke_cleaners2"],
"max_wav_value": 32768.0,
"sampling_rate": 22050,
"filter_length": 1024,
"hop_length": 256,
"win_length": 1024,
"n_mel_channels": 80,
"mel_fmin": 0.0,
"mel_fmax": null,
"add_blank": true,
"n_speakers": 2891,
"cleaned_text": true
},
"model": {
"inter_channels": 192,
"hidden_channels": 192,
"filter_channels": 768,
"n_heads": 2,
"n_layers": 6,
"kernel_size": 3,
"p_dropout": 0.1,
"resblock": "1",
"resblock_kernel_sizes": [3,7,11],
"resblock_dilation_sizes": [[1,3,5], [1,3,5], [1,3,5]],
"upsample_rates": [8,8,2,2],
"upsample_initial_channel": 512,
"upsample_kernel_sizes": [16,16,4,4],
"n_layers_q": 3,
"use_spectral_norm": false,
"gin_channels": 256
},
"symbols": ["_", ",", ".", "!", "?", "-", "~", "\u2026", "N", "Q", "a", "b", "d", "e", "f", "g", "h", "i", "j", "k", "l", "m", "n", "o", "p", "s", "t", "u", "v", "w", "x", "y", "z", "\u0251", "\u00e6", "\u0283", "\u0291", "\u00e7", "\u026f", "\u026a", "\u0254", "\u025b", "\u0279", "\u00f0", "\u0259", "\u026b", "\u0265", "\u0278", "\u028a", "\u027e", "\u0292", "\u03b8", "\u03b2", "\u014b", "\u0266", "\u207c", "\u02b0", "`", "^", "#", "*", "=", "\u02c8", "\u02cc", "\u2192", "\u2193", "\u2191", " "]
}
configs/cjks_base.json
-55
View File
@@ -1,55 +0,0 @@
{
"train": {
"log_interval": 200,
"eval_interval": 1000,
"seed": 1234,
"epochs": 10000,
"learning_rate": 2e-4,
"betas": [0.8, 0.99],
"eps": 1e-9,
"batch_size": 32,
"fp16_run": true,
"lr_decay": 0.999875,
"segment_size": 8192,
"init_lr_ratio": 1,
"warmup_epochs": 0,
"c_mel": 45,
"c_kl": 1.0
},
"data": {
"training_files":"filelists/cjks_train_filelist.txt.cleaned",
"validation_files":"filelists/cjks_val_filelist.txt.cleaned",
"text_cleaners":["cjks_cleaners"],
"max_wav_value": 32768.0,
"sampling_rate": 22050,
"filter_length": 1024,
"hop_length": 256,
"win_length": 1024,
"n_mel_channels": 80,
"mel_fmin": 0.0,
"mel_fmax": null,
"add_blank": true,
"n_speakers": 24,
"cleaned_text": true
},
"model": {
"inter_channels": 192,
"hidden_channels": 192,
"filter_channels": 768,
"n_heads": 2,
"n_layers": 6,
"kernel_size": 3,
"p_dropout": 0.1,
"resblock": "1",
"resblock_kernel_sizes": [3,7,11],
"resblock_dilation_sizes": [[1,3,5], [1,3,5], [1,3,5]],
"upsample_rates": [8,8,2,2],
"upsample_initial_channel": 512,
"upsample_kernel_sizes": [16,16,4,4],
"n_layers_q": 3,
"use_spectral_norm": false,
"gin_channels": 256
},
"speakers": ["\u7dbe\u5730\u5be7\u3005", "\u671d\u6b66\u82b3\u4e43", "\u5728\u539f\u4e03\u6d77", "\u30eb\u30a4\u30ba", "\u91d1\u8272\u306e\u95c7", "\u30e2\u30e2", "\u7d50\u57ce\u7f8e\u67d1", "\u5c0f\u8338", "\u5510\u4e50\u541f", "\u5c0f\u6bb7", "\u82b1\u73b2", "\u516b\u56db", "\uc218\uc544", "\ubbf8\ubbf8\ub974", "\uc544\ub9b0", "\uc720\ud654", "\uc5f0\ud654", "SA1", "SA2", "SA3", "SA4", "SA5", "SA6", ""],
"symbols": ["_", ",", ".", "!", "?", "-", "~", "\u2026", "N", "Q", "a", "b", "d", "e", "f", "g", "h", "i", "j", "k", "l", "m", "n", "o", "p", "s", "t", "u", "v", "w", "x", "y", "z", "\u0283", "\u02a7", "\u02a5", "\u02a6", "\u026f", "\u0279", "\u0259", "\u0265", "\u00e7", "\u0278", "\u027e", "\u03b2", "\u014b", "\u0266", "\u02d0", "\u207c", "\u02b0", "`", "^", "#", "*", "=", "\u2192", "\u2193", "\u2191", " "]
}
configs/finetune_speaker.json
+2 -2
View File
@@ -17,8 +17,8 @@
"c_kl": 1.0
},
"data": {
"training_files":"../CH_JA_EN_mix_voice/rosalia/rosalia.txt.cleaned",
"validation_files":"../CH_JA_EN_mix_voice/rosalia/rosalia.txt.cleaned",
"training_files":"final_annotation.txt",
"validation_files":"final_annotation.txt",
"text_cleaners":["cjke_cleaners2"],
"max_wav_value": 32768.0,
"sampling_rate": 22050,
configs/japanese_base.json
-55
View File
@@ -1,55 +0,0 @@
{
"train": {
"log_interval": 200,
"eval_interval": 1000,
"seed": 1234,
"epochs": 10000,
"learning_rate": 2e-4,
"betas": [0.8, 0.99],
"eps": 1e-9,
"batch_size": 32,
"fp16_run": true,
"lr_decay": 0.999875,
"segment_size": 8192,
"init_lr_ratio": 1,
"warmup_epochs": 0,
"c_mel": 45,
"c_kl": 1.0
},
"data": {
"training_files":"filelists/train_filelist.txt.cleaned",
"validation_files":"filelists/val_filelist.txt.cleaned",
"text_cleaners":["japanese_cleaners"],
"max_wav_value": 32768.0,
"sampling_rate": 22050,
"filter_length": 1024,
"hop_length": 256,
"win_length": 1024,
"n_mel_channels": 80,
"mel_fmin": 0.0,
"mel_fmax": null,
"add_blank": true,
"n_speakers": 7,
"cleaned_text": true
},
"model": {
"inter_channels": 192,
"hidden_channels": 192,
"filter_channels": 768,
"n_heads": 2,
"n_layers": 6,
"kernel_size": 3,
"p_dropout": 0.1,
"resblock": "1",
"resblock_kernel_sizes": [3,7,11],
"resblock_dilation_sizes": [[1,3,5], [1,3,5], [1,3,5]],
"upsample_rates": [8,8,2,2],
"upsample_initial_channel": 512,
"upsample_kernel_sizes": [16,16,4,4],
"n_layers_q": 3,
"use_spectral_norm": false,
"gin_channels": 256
},
"speakers": ["\u7dbe\u5730\u5be7\u3005", "\u56e0\u5e61\u3081\u3050\u308b", "\u671d\u6b66\u82b3\u4e43", "\u5e38\u9678\u8309\u5b50", "\u30e0\u30e9\u30b5\u30e1", "\u978d\u99ac\u5c0f\u6625", "\u5728\u539f\u4e03\u6d77"],
"symbols": ["_", ",", ".", "!", "?", "-", "A", "E", "I", "N", "O", "Q", "U", "a", "b", "d", "e", "f", "g", "h", "i", "j", "k", "m", "n", "o", "p", "r", "s", "t", "u", "v", "w", "y", "z", "\u0283", "\u02a7", "\u2193", "\u2191", " "]
}
configs/japanese_base2.json
-55
View File
@@ -1,55 +0,0 @@
{
"train": {
"log_interval": 200,
"eval_interval": 1000,
"seed": 1234,
"epochs": 10000,
"learning_rate": 2e-4,
"betas": [0.8, 0.99],
"eps": 1e-9,
"batch_size": 32,
"fp16_run": true,
"lr_decay": 0.999875,
"segment_size": 8192,
"init_lr_ratio": 1,
"warmup_epochs": 0,
"c_mel": 45,
"c_kl": 1.0
},
"data": {
"training_files":"filelists/hamidashi_train_filelist.txt.cleaned",
"validation_files":"filelists/hamidashi_val_filelist.txt.cleaned",
"text_cleaners":["japanese_cleaners2"],
"max_wav_value": 32768.0,
"sampling_rate": 22050,
"filter_length": 1024,
"hop_length": 256,
"win_length": 1024,
"n_mel_channels": 80,
"mel_fmin": 0.0,
"mel_fmax": null,
"add_blank": true,
"n_speakers": 8,
"cleaned_text": true
},
"model": {
"inter_channels": 192,
"hidden_channels": 192,
"filter_channels": 768,
"n_heads": 2,
"n_layers": 6,
"kernel_size": 3,
"p_dropout": 0.1,
"resblock": "1",
"resblock_kernel_sizes": [3,7,11],
"resblock_dilation_sizes": [[1,3,5], [1,3,5], [1,3,5]],
"upsample_rates": [8,8,2,2],
"upsample_initial_channel": 512,
"upsample_kernel_sizes": [16,16,4,4],
"n_layers_q": 3,
"use_spectral_norm": false,
"gin_channels": 256
},
"speakers": ["\u548c\u6cc9\u5983\u611b", "\u5e38\u76e4\u83ef\u4e43", "\u9326\u3042\u3059\u307f", "\u938c\u5009\u8a69\u685c", "\u7adc\u9591\u5929\u68a8", "\u548c\u6cc9\u91cc", "\u65b0\u5ddd\u5e83\u5922", "\u8056\u8389\u3005\u5b50"],
"symbols": ["_", ",", ".", "!", "?", "-", "~", "\u2026", "A", "E", "I", "N", "O", "Q", "U", "a", "b", "d", "e", "f", "g", "h", "i", "j", "k", "m", "n", "o", "p", "r", "s", "t", "u", "v", "w", "y", "z", "\u0283", "\u02a7", "\u02a6", "\u2193", "\u2191", " "]
}
configs/japanese_ss_base2.json
-54
View File
@@ -1,54 +0,0 @@
{
"train": {
"log_interval": 200,
"eval_interval": 1000,
"seed": 1234,
"epochs": 20000,
"learning_rate": 2e-4,
"betas": [0.8, 0.99],
"eps": 1e-9,
"batch_size": 32,
"fp16_run": true,
"lr_decay": 0.999875,
"segment_size": 8192,
"init_lr_ratio": 1,
"warmup_epochs": 0,
"c_mel": 45,
"c_kl": 1.0
},
"data": {
"training_files":"filelists/train_filelist.txt.cleaned",
"validation_files":"filelists/val_filelist.txt.cleaned",
"text_cleaners":["japanese_cleaners2"],
"max_wav_value": 32768.0,
"sampling_rate": 22050,
"filter_length": 1024,
"hop_length": 256,
"win_length": 1024,
"n_mel_channels": 80,
"mel_fmin": 0.0,
"mel_fmax": null,
"add_blank": true,
"n_speakers": 0,
"cleaned_text": true
},
"model": {
"inter_channels": 192,
"hidden_channels": 192,
"filter_channels": 768,
"n_heads": 2,
"n_layers": 6,
"kernel_size": 3,
"p_dropout": 0.1,
"resblock": "1",
"resblock_kernel_sizes": [3,7,11],
"resblock_dilation_sizes": [[1,3,5], [1,3,5], [1,3,5]],
"upsample_rates": [8,8,2,2],
"upsample_initial_channel": 512,
"upsample_kernel_sizes": [16,16,4,4],
"n_layers_q": 3,
"use_spectral_norm": false
},
"speakers": ["\u30eb\u30a4\u30ba"],
"symbols": ["_", ",", ".", "!", "?", "-", "~", "\u2026", "A", "E", "I", "N", "O", "Q", "U", "a", "b", "d", "e", "f", "g", "h", "i", "j", "k", "m", "n", "o", "p", "r", "s", "t", "u", "v", "w", "y", "z", "\u0283", "\u02a7", "\u02a6", "\u2193", "\u2191", " "]
}
configs/korean_base.json
-55
View File
@@ -1,55 +0,0 @@
{
"train": {
"log_interval": 200,
"eval_interval": 1000,
"seed": 1234,
"epochs": 10000,
"learning_rate": 2e-4,
"betas": [0.8, 0.99],
"eps": 1e-9,
"batch_size": 32,
"fp16_run": true,
"lr_decay": 0.999875,
"segment_size": 8192,
"init_lr_ratio": 1,
"warmup_epochs": 0,
"c_mel": 45,
"c_kl": 1.0
},
"data": {
"training_files":"filelists/fox_train_filelist.txt.cleaned",
"validation_files":"filelists/fox_val_filelist.txt.cleaned",
"text_cleaners":["korean_cleaners"],
"max_wav_value": 32768.0,
"sampling_rate": 22050,
"filter_length": 1024,
"hop_length": 256,
"win_length": 1024,
"n_mel_channels": 80,
"mel_fmin": 0.0,
"mel_fmax": null,
"add_blank": true,
"n_speakers": 6,
"cleaned_text": true
},
"model": {
"inter_channels": 192,
"hidden_channels": 192,
"filter_channels": 768,
"n_heads": 2,
"n_layers": 6,
"kernel_size": 3,
"p_dropout": 0.1,
"resblock": "1",
"resblock_kernel_sizes": [3,7,11],
"resblock_dilation_sizes": [[1,3,5], [1,3,5], [1,3,5]],
"upsample_rates": [8,8,2,2],
"upsample_initial_channel": 512,
"upsample_kernel_sizes": [16,16,4,4],
"n_layers_q": 3,
"use_spectral_norm": false,
"gin_channels": 256
},
"speakers": ["\uc218\uc544", "\ubbf8\ubbf8\ub974", "\uc544\ub9b0", "\uc5f0\ud654", "\uc720\ud654", "\uc120\ubc30"],
"symbols": ["_", ",", ".", "!", "?", "\u2026", "~", "\u3131", "\u3134", "\u3137", "\u3139", "\u3141", "\u3142", "\u3145", "\u3147", "\u3148", "\u314a", "\u314b", "\u314c", "\u314d", "\u314e", "\u3132", "\u3138", "\u3143", "\u3146", "\u3149", "\u314f", "\u3153", "\u3157", "\u315c", "\u3161", "\u3163", "\u3150", "\u3154", " "]
}
configs/sanskrit_base.json
-55
View File
@@ -1,55 +0,0 @@
{
"train": {
"log_interval": 200,
"eval_interval": 1000,
"seed": 1234,
"epochs": 10000,
"learning_rate": 2e-4,
"betas": [0.8, 0.99],
"eps": 1e-9,
"batch_size": 32,
"fp16_run": true,
"lr_decay": 0.999875,
"segment_size": 8192,
"init_lr_ratio": 1,
"warmup_epochs": 0,
"c_mel": 45,
"c_kl": 1.0
},
"data": {
"training_files":"filelists/sanskrit_train_filelist.txt.cleaned",
"validation_files":"filelists/sanskrit_val_filelist.txt.cleaned",
"text_cleaners":["sanskrit_cleaners"],
"max_wav_value": 32768.0,
"sampling_rate": 22050,
"filter_length": 1024,
"hop_length": 256,
"win_length": 1024,
"n_mel_channels": 80,
"mel_fmin": 0.0,
"mel_fmax": null,
"add_blank": true,
"n_speakers": 27,
"cleaned_text": true
},
"model": {
"inter_channels": 192,
"hidden_channels": 192,
"filter_channels": 768,
"n_heads": 2,
"n_layers": 6,
"kernel_size": 3,
"p_dropout": 0.1,
"resblock": "1",
"resblock_kernel_sizes": [3,7,11],
"resblock_dilation_sizes": [[1,3,5], [1,3,5], [1,3,5]],
"upsample_rates": [8,8,2,2],
"upsample_initial_channel": 512,
"upsample_kernel_sizes": [16,16,4,4],
"n_layers_q": 3,
"use_spectral_norm": false,
"gin_channels": 256
},
"speakers": ["Male 1", "Male 2", "Male 3", "Male 4 (Malayalam)", "Male 5", "Male 6", "Male 7", "Male 8 (Kannada)", "Female 1 (Tamil)", "Male 9 (Kannada)", "Female 2 (Marathi)", "Female 3 (Marathi)", "Female 4 (Marathi)", "Female 5 (Telugu)", "Female 6 (Telugu)", "Male 10 (Kannada)", "Male 11 (Kannada)", "Male 12", "Male 13", "Male 14", "Male 15", "Female 7", "Male 16 (Malayalam)", "Male 17 (Tamil)", "Male 18 (Hindi)", "Male 19 (Telugu)", "Male 20 (Hindi)"],
"symbols": ["_", "\u0964", "\u0901", "\u0902", "\u0903", "\u0905", "\u0906", "\u0907", "\u0908", "\u0909", "\u090a", "\u090b", "\u090f", "\u0910", "\u0913", "\u0914", "\u0915", "\u0916", "\u0917", "\u0918", "\u0919", "\u091a", "\u091b", "\u091c", "\u091d", "\u091e", "\u091f", "\u0920", "\u0921", "\u0922", "\u0923", "\u0924", "\u0925", "\u0926", "\u0927", "\u0928", "\u092a", "\u092b", "\u092c", "\u092d", "\u092e", "\u092f", "\u0930", "\u0932", "\u0933", "\u0935", "\u0936", "\u0937", "\u0938", "\u0939", "\u093d", "\u093e", "\u093f", "\u0940", "\u0941", "\u0942", "\u0943", "\u0944", "\u0947", "\u0948", "\u094b", "\u094c", "\u094d", "\u0960", "\u0962", " "]
}
configs/shanghainese_base.json
-55
View File
@@ -1,55 +0,0 @@
{
"train": {
"log_interval": 200,
"eval_interval": 1000,
"seed": 1234,
"epochs": 10000,
"learning_rate": 2e-4,
"betas": [0.8, 0.99],
"eps": 1e-9,
"batch_size": 32,
"fp16_run": true,
"lr_decay": 0.999875,
"segment_size": 8192,
"init_lr_ratio": 1,
"warmup_epochs": 0,
"c_mel": 45,
"c_kl": 1.0
},
"data": {
"training_files":"filelists/zaonhe_train_filelist.txt.cleaned",
"validation_files":"filelists/zaonhe_val_filelist.txt.cleaned",
"text_cleaners":["shanghainese_cleaners"],
"max_wav_value": 32768.0,
"sampling_rate": 22050,
"filter_length": 1024,
"hop_length": 256,
"win_length": 1024,
"n_mel_channels": 80,
"mel_fmin": 0.0,
"mel_fmax": null,
"add_blank": true,
"n_speakers": 2,
"cleaned_text": true
},
"model": {
"inter_channels": 192,
"hidden_channels": 192,
"filter_channels": 768,
"n_heads": 2,
"n_layers": 6,
"kernel_size": 3,
"p_dropout": 0.1,
"resblock": "1",
"resblock_kernel_sizes": [3,7,11],
"resblock_dilation_sizes": [[1,3,5], [1,3,5], [1,3,5]],
"upsample_rates": [8,8,2,2],
"upsample_initial_channel": 512,
"upsample_kernel_sizes": [16,16,4,4],
"n_layers_q": 3,
"use_spectral_norm": false,
"gin_channels": 256
},
"speakers": ["1", "2"],
"symbols": ["_", ",", ".", "!", "?", "\u2026", "a", "b", "d", "f", "g", "h", "i", "k", "l", "m", "n", "o", "p", "s", "t", "u", "v", "y", "z", "\u00f8", "\u014b", "\u0235", "\u0251", "\u0254", "\u0255", "\u0259", "\u0264", "\u0266", "\u026a", "\u027f", "\u0291", "\u0294", "\u02b0", "\u0303", "\u0329", "\u1d00", "\u1d07", "1", "5", "6", "7", "8", " "]
}
configs/uma87.json
-142
View File
@@ -1,142 +0,0 @@
{
"train": {
"log_interval": 200,
"eval_interval": 1000,
"seed": 1234,
"epochs": 10000,
"learning_rate": 2e-4,
"betas": [0.8, 0.99],
"eps": 1e-9,
"batch_size": 1,
"fp16_run": true,
"lr_decay": 0.999875,
"segment_size": 8192,
"init_lr_ratio": 1,
"warmup_epochs": 0,
"c_mel": 45,
"c_kl": 1.0
},
"data": {
"training_files":"E:/uma_voice/output_train.txt.cleaned",
"validation_files":"E:/uma_voice/output_val.txt.cleaned",
"text_cleaners":["japanese_cleaners"],
"max_wav_value": 32768.0,
"sampling_rate": 22050,
"filter_length": 1024,
"hop_length": 256,
"win_length": 1024,
"n_mel_channels": 80,
"mel_fmin": 0.0,
"mel_fmax": null,
"add_blank": true,
"n_speakers": 87,
"cleaned_text": true
},
"model": {
"inter_channels": 192,
"hidden_channels": 192,
"filter_channels": 768,
"n_heads": 2,
"n_layers": 6,
"kernel_size": 3,
"p_dropout": 0.1,
"resblock": "1",
"resblock_kernel_sizes": [3,7,11],
"resblock_dilation_sizes": [[1,3,5], [1,3,5], [1,3,5]],
"upsample_rates": [8,8,2,2],
"upsample_initial_channel": 512,
"upsample_kernel_sizes": [16,16,4,4],
"n_layers_q": 3,
"use_spectral_norm": false,
"gin_channels": 256
},
"speakers": ["Special Week",
"Silence Suzuka",
"Tokai Teio",
"Maruzensky",
"Fuji Kiseki",
"Oguri Cap",
"Gold Ship",
"Vodka",
"Daiwa Scarlet",
"Taiki Shuttle",
"Grass Wonder",
"Hishi Amazon",
"Mejiro Mcqueen",
"El Condor Pasa",
"T.M. Opera O",
"Narita Brian",
"Symboli Rudolf",
"Air Groove",
"Agnes Digital",
"Seiun Sky",
"Tamamo Cross",
"Fine Motion",
"Biwa Hayahide",
"Mayano Topgun",
"Manhattan Cafe",
"Mihono Bourbon",
"Mejiro Ryan",
"Hishi Akebono",
"Yukino Bijin",
"Rice Shower",
"Ines Fujin",
"Agnes Tachyon",
"Admire Vega",
"Inari One",
"Winning Ticket",
"Air Shakur",
"Eishin Flash",
"Curren Chan",
"Kawakami Princess",
"Gold City",
"Sakura Bakushin O",
"Seeking the Pearl",
"Shinko Windy",
"Sweep Tosho",
"Super Creek",
"Smart Falcon",
"Zenno Rob Roy",
"Tosen Jordan",
"Nakayama Festa",
"Narita Taishin",
"Nishino Flower",
"Haru Urara",
"Bamboo Memory",
"Biko Pegasus",
"Marvelous Sunday",
"Matikane Fukukitaru",
"Mr. C.B.",
"Meisho Doto",
"Mejiro Dober",
"Nice Nature",
"King Halo",
"Matikane Tannhauser",
"Ikuno Dictus",
"Mejiro Palmer",
"Daitaku Helios",
"Twin Turbo",
"Satono Diamond",
"Kitasan Black",
"Sakura Chiyono O",
"Sirius Symboli",
"Mejiro Ardan",
"Yaeno Muteki",
"Tsurumaru Tsuyoshi",
"Mejiro Bright",
"Sakura Laurel",
"Narita Top Road",
"Yamanin Zephyr",
"Symboli Kris S",
"Tanino Gimlet",
"Daiichi Ruby",
"Aston Machan",
"Hayakawa Tazuna",
"KS Miracle",
"Kopano Rickey",
"Hoko Tarumae",
"Wonder Acute",
"President Akikawa"
],
"symbols": ["_", ",", ".", "!", "?", "-", "A", "E", "I", "N", "O", "Q", "U", "a", "b", "d", "e", "f", "g", "h", "i", "j", "k", "m", "n", "o", "p", "r", "s", "t", "u", "v", "w", "y", "z", "\u0283", "\u02a7", "\u2193", "\u2191", " "]
}
configs/zero_japanese_base2.json
-55
View File
@@ -1,55 +0,0 @@
{
"train": {
"log_interval": 200,
"eval_interval": 1000,
"seed": 1234,
"epochs": 10000,
"learning_rate": 2e-4,
"betas": [0.8, 0.99],
"eps": 1e-9,
"batch_size": 32,
"fp16_run": true,
"lr_decay": 0.999875,
"segment_size": 8192,
"init_lr_ratio": 1,
"warmup_epochs": 0,
"c_mel": 45,
"c_kl": 1.0
},
"data": {
"training_files":"filelists/zero_train_filelist.txt.cleaned",
"validation_files":"filelists/zero_val_filelist.txt.cleaned",
"text_cleaners":["japanese_cleaners2"],
"max_wav_value": 32768.0,
"sampling_rate": 22050,
"filter_length": 1024,
"hop_length": 256,
"win_length": 1024,
"n_mel_channels": 80,
"mel_fmin": 0.0,
"mel_fmax": null,
"add_blank": true,
"n_speakers": 26,
"cleaned_text": true
},
"model": {
"inter_channels": 192,
"hidden_channels": 192,
"filter_channels": 768,
"n_heads": 2,
"n_layers": 6,
"kernel_size": 3,
"p_dropout": 0.1,
"resblock": "1",
"resblock_kernel_sizes": [3,7,11],
"resblock_dilation_sizes": [[1,3,5], [1,3,5], [1,3,5]],
"upsample_rates": [8,8,2,2],
"upsample_initial_channel": 512,
"upsample_kernel_sizes": [16,16,4,4],
"n_layers_q": 3,
"use_spectral_norm": false,
"gin_channels": 256
},
"speakers": ["\u30eb\u30a4\u30ba", "\u30c6\u30a3\u30d5\u30a1\u30cb\u30a2", "\u30a4\u30eb\u30af\u30af\u30a5", "\u30a2\u30f3\u30ea\u30a8\u30c3\u30bf", "\u30bf\u30d0\u30b5", "\u30b7\u30a8\u30b9\u30bf", "\u30cf\u30eb\u30ca", "\u5c11\u5973\u30ea\u30b7\u30e5", "\u30ea\u30b7\u30e5", "\u30a2\u30ad\u30ca", "\u30af\u30ea\u30b9", "\u30ab\u30c8\u30ec\u30a2", "\u30a8\u30ec\u30aa\u30ce\u30fc\u30eb", "\u30e2\u30f3\u30e2\u30e9\u30f3\u30b7\u30fc", "\u30ea\u30fc\u30f4\u30eb", "\u30ad\u30e5\u30eb\u30b1", "\u30a6\u30a7\u30b6\u30ea\u30fc", "\u30b5\u30a4\u30c8", "\u30ae\u30fc\u30b7\u30e5", "\u30b3\u30eb\u30d9\u30fc\u30eb", "\u30aa\u30b9\u30de\u30f3", "\u30c7\u30eb\u30d5\u30ea\u30f3\u30ac\u30fc", "\u30c6\u30af\u30b9\u30c8", "\u30c0\u30f3\u30d7\u30ea\u30e1", "\u30ac\u30ec\u30c3\u30c8", "\u30b9\u30ab\u30ed\u30f3"],
"symbols": ["_", ",", ".", "!", "?", "-", "~", "\u2026", "A", "E", "I", "N", "O", "Q", "U", "a", "b", "d", "e", "f", "g", "h", "i", "j", "k", "m", "n", "o", "p", "r", "s", "t", "u", "v", "w", "y", "z", "\u0283", "\u02a7", "\u02a6", "\u2193", "\u2191", " "]
}
configs/zh_ja_mixture_base.json
-55
View File
@@ -1,55 +0,0 @@
{
"train": {
"log_interval": 200,
"eval_interval": 1000,
"seed": 1234,
"epochs": 10000,
"learning_rate": 2e-4,
"betas": [0.8, 0.99],
"eps": 1e-9,
"batch_size": 32,
"fp16_run": true,
"lr_decay": 0.999875,
"segment_size": 8192,
"init_lr_ratio": 1,
"warmup_epochs": 0,
"c_mel": 45,
"c_kl": 1.0
},
"data": {
"training_files":"filelists/mix_train_filelist.txt.cleaned",
"validation_files":"filelists/mix_val_filelist.txt.cleaned",
"text_cleaners":["zh_ja_mixture_cleaners"],
"max_wav_value": 32768.0,
"sampling_rate": 22050,
"filter_length": 1024,
"hop_length": 256,
"win_length": 1024,
"n_mel_channels": 80,
"mel_fmin": 0.0,
"mel_fmax": null,
"add_blank": true,
"n_speakers": 5,
"cleaned_text": true
},
"model": {
"inter_channels": 192,
"hidden_channels": 192,
"filter_channels": 768,
"n_heads": 2,
"n_layers": 6,
"kernel_size": 3,
"p_dropout": 0.1,
"resblock": "1",
"resblock_kernel_sizes": [3,7,11],
"resblock_dilation_sizes": [[1,3,5], [1,3,5], [1,3,5]],
"upsample_rates": [8,8,2,2],
"upsample_initial_channel": 512,
"upsample_kernel_sizes": [16,16,4,4],
"n_layers_q": 3,
"use_spectral_norm": false,
"gin_channels": 256
},
"speakers": ["\u7dbe\u5730\u5be7\u3005", "\u5728\u539f\u4e03\u6d77", "\u5c0f\u8338", "\u5510\u4e50\u541f"],
"symbols": ["_", ",", ".", "!", "?", "-", "~", "\u2026", "A", "E", "I", "N", "O", "Q", "U", "a", "b", "d", "e", "f", "g", "h", "i", "j", "k", "l", "m", "n", "o", "p", "r", "s", "t", "u", "v", "w", "y", "z", "\u0283", "\u02a7", "\u02a6", "\u026f", "\u0279", "\u0259", "\u0265", "\u207c", "\u02b0", "`", "\u2192", "\u2193", "\u2191", " "]
}
finetune_speaker.py
+319
View File
@@ -0,0 +1,319 @@
import os
import json
import argparse
import itertools
import math
import torch
from torch import nn, optim
from torch.nn import functional as F
from torch.utils.data import DataLoader
from torch.utils.tensorboard import SummaryWriter
import torch.multiprocessing as mp
import torch.distributed as dist
from torch.nn.parallel import DistributedDataParallel as DDP
from torch.cuda.amp import autocast, GradScaler
from tqdm import tqdm
import librosa
import logging
logging.getLogger('numba').setLevel(logging.WARNING)
import commons
import utils
from data_utils import (
TextAudioSpeakerLoader,
TextAudioSpeakerCollate,
DistributedBucketSampler
)
from models import (
SynthesizerTrn,
MultiPeriodDiscriminator,
)
from losses import (
generator_loss,
discriminator_loss,
feature_loss,
kl_loss
)
from mel_processing import mel_spectrogram_torch, spec_to_mel_torch
torch.backends.cudnn.benchmark = True
global_step = 0
def main():
"""Assume Single Node Multi GPUs Training Only"""
assert torch.cuda.is_available(), "CPU training is not allowed."
n_gpus = torch.cuda.device_count()
os.environ['MASTER_ADDR'] = 'localhost'
os.environ['MASTER_PORT'] = '8000'
hps = utils.get_hparams()
mp.spawn(run, nprocs=n_gpus, args=(n_gpus, hps,))
def run(rank, n_gpus, hps):
global global_step
symbols = hps['symbols']
if rank == 0:
logger = utils.get_logger(hps.model_dir)
logger.info(hps)
utils.check_git_hash(hps.model_dir)
writer = SummaryWriter(log_dir=hps.model_dir)
writer_eval = SummaryWriter(log_dir=os.path.join(hps.model_dir, "eval"))
dist.init_process_group(backend='nccl', init_method='env://', world_size=n_gpus, rank=rank)
torch.manual_seed(hps.train.seed)
torch.cuda.set_device(rank)
train_dataset = TextAudioSpeakerLoader(hps.data.training_files, hps.data)
train_sampler = DistributedBucketSampler(
train_dataset,
hps.train.batch_size,
[32,300,400,500,600,700,800,900,1000],
num_replicas=n_gpus,
rank=rank,
shuffle=True)
collate_fn = TextAudioSpeakerCollate()
train_loader = DataLoader(train_dataset, num_workers=0, shuffle=False, pin_memory=True,
collate_fn=collate_fn, batch_sampler=train_sampler)
# train_loader = DataLoader(train_dataset, batch_size=hps.train.batch_size, num_workers=0, shuffle=False, pin_memory=True,
# collate_fn=collate_fn)
if rank == 0:
eval_dataset = TextAudioSpeakerLoader(hps.data.validation_files, hps.data)
eval_loader = DataLoader(eval_dataset, num_workers=0, shuffle=False,
batch_size=hps.train.batch_size, pin_memory=True,
drop_last=False, collate_fn=collate_fn)
net_g = SynthesizerTrn(
len(symbols),
hps.data.filter_length // 2 + 1,
hps.train.segment_size // hps.data.hop_length,
n_speakers=hps.data.n_speakers,
**hps.model).cuda(rank)
net_d = MultiPeriodDiscriminator(hps.model.use_spectral_norm).cuda(rank)
# load existing model
_, _, _, _ = utils.load_checkpoint("./pretrained_models/G_trilingual.pth", net_g, None)
_, _, _, _ = utils.load_checkpoint("./pretrained_models/D_trilingual.pth", net_d, None)
epoch_str = 1
global_step = 0
# freeze all other layers except speaker embedding
for p in net_g.parameters():
p.requires_grad = True
for p in net_d.parameters():
p.requires_grad = True
# for p in net_d.parameters():
# p.requires_grad = False
# net_g.emb_g.weight.requires_grad = True
optim_g = torch.optim.AdamW(
net_g.parameters(),
hps.train.learning_rate,
betas=hps.train.betas,
eps=hps.train.eps)
optim_d = torch.optim.AdamW(
net_d.parameters(),
hps.train.learning_rate,
betas=hps.train.betas,
eps=hps.train.eps)
# optim_d = None
net_g = DDP(net_g, device_ids=[rank])
net_d = DDP(net_d, device_ids=[rank])
scheduler_g = torch.optim.lr_scheduler.ExponentialLR(optim_g, gamma=hps.train.lr_decay)
scheduler_d = torch.optim.lr_scheduler.ExponentialLR(optim_d, gamma=hps.train.lr_decay)
scaler = GradScaler(enabled=hps.train.fp16_run)
for epoch in range(epoch_str, hps.train.epochs + 1):
if rank==0:
train_and_evaluate(rank, epoch, hps, [net_g, net_d], [optim_g, optim_d], [scheduler_g, scheduler_d], scaler, [train_loader, eval_loader], logger, [writer, writer_eval])
else:
train_and_evaluate(rank, epoch, hps, [net_g, net_d], [optim_g, optim_d], [scheduler_g, scheduler_d], scaler, [train_loader, None], None, None)
scheduler_g.step()
scheduler_d.step()
def train_and_evaluate(rank, epoch, hps, nets, optims, schedulers, scaler, loaders, logger, writers):
net_g, net_d = nets
optim_g, optim_d = optims
scheduler_g, scheduler_d = schedulers
train_loader, eval_loader = loaders
if writers is not None:
writer, writer_eval = writers
# train_loader.batch_sampler.set_epoch(epoch)
global global_step
net_g.train()
net_d.train()
for batch_idx, (x, x_lengths, spec, spec_lengths, y, y_lengths, speakers) in enumerate(tqdm(train_loader)):
x, x_lengths = x.cuda(rank, non_blocking=True), x_lengths.cuda(rank, non_blocking=True)
spec, spec_lengths = spec.cuda(rank, non_blocking=True), spec_lengths.cuda(rank, non_blocking=True)
y, y_lengths = y.cuda(rank, non_blocking=True), y_lengths.cuda(rank, non_blocking=True)
speakers = speakers.cuda(rank, non_blocking=True)
with autocast(enabled=hps.train.fp16_run):
y_hat, l_length, attn, ids_slice, x_mask, z_mask,\
(z, z_p, m_p, logs_p, m_q, logs_q) = net_g(x, x_lengths, spec, spec_lengths, speakers)
mel = spec_to_mel_torch(
spec,
hps.data.filter_length,
hps.data.n_mel_channels,
hps.data.sampling_rate,
hps.data.mel_fmin,
hps.data.mel_fmax)
y_mel = commons.slice_segments(mel, ids_slice, hps.train.segment_size // hps.data.hop_length)
y_hat_mel = mel_spectrogram_torch(
y_hat.squeeze(1),
hps.data.filter_length,
hps.data.n_mel_channels,
hps.data.sampling_rate,
hps.data.hop_length,
hps.data.win_length,
hps.data.mel_fmin,
hps.data.mel_fmax
)
y = commons.slice_segments(y, ids_slice * hps.data.hop_length, hps.train.segment_size) # slice
# Discriminator
y_d_hat_r, y_d_hat_g, _, _ = net_d(y, y_hat.detach())
with autocast(enabled=False):
loss_disc, losses_disc_r, losses_disc_g = discriminator_loss(y_d_hat_r, y_d_hat_g)
loss_disc_all = loss_disc
optim_d.zero_grad()
scaler.scale(loss_disc_all).backward()
scaler.unscale_(optim_d)
grad_norm_d = commons.clip_grad_value_(net_d.parameters(), None)
scaler.step(optim_d)
with autocast(enabled=hps.train.fp16_run):
# Generator
y_d_hat_r, y_d_hat_g, fmap_r, fmap_g = net_d(y, y_hat)
with autocast(enabled=False):
loss_dur = torch.sum(l_length.float())
loss_mel = F.l1_loss(y_mel, y_hat_mel) * hps.train.c_mel
loss_kl = kl_loss(z_p, logs_q, m_p, logs_p, z_mask) * hps.train.c_kl
loss_fm = feature_loss(fmap_r, fmap_g)
loss_gen, losses_gen = generator_loss(y_d_hat_g)
loss_gen_all = loss_gen + loss_fm + loss_mel + loss_dur + loss_kl
optim_g.zero_grad()
scaler.scale(loss_gen_all).backward()
scaler.unscale_(optim_g)
grad_norm_g = commons.clip_grad_value_(net_g.parameters(), None)
scaler.step(optim_g)
scaler.update()
if rank==0:
if global_step % hps.train.log_interval == 0:
lr = optim_g.param_groups[0]['lr']
losses = [loss_disc, loss_gen, loss_fm, loss_mel, loss_dur, loss_kl]
logger.info('Train Epoch: {} [{:.0f}%]'.format(
epoch,
100. * batch_idx / len(train_loader)))
logger.info([x.item() for x in losses] + [global_step, lr])
scalar_dict = {"loss/g/total": loss_gen_all, "loss/d/total": loss_disc_all, "learning_rate": lr, "grad_norm_g": grad_norm_g}
scalar_dict.update({"loss/g/fm": loss_fm, "loss/g/mel": loss_mel, "loss/g/dur": loss_dur, "loss/g/kl": loss_kl})
scalar_dict.update({"loss/g/{}".format(i): v for i, v in enumerate(losses_gen)})
scalar_dict.update({"loss/d_r/{}".format(i): v for i, v in enumerate(losses_disc_r)})
scalar_dict.update({"loss/d_g/{}".format(i): v for i, v in enumerate(losses_disc_g)})
image_dict = {
"slice/mel_org": utils.plot_spectrogram_to_numpy(y_mel[0].data.cpu().numpy()),
"slice/mel_gen": utils.plot_spectrogram_to_numpy(y_hat_mel[0].data.cpu().numpy()),
"all/mel": utils.plot_spectrogram_to_numpy(mel[0].data.cpu().numpy()),
"all/attn": utils.plot_alignment_to_numpy(attn[0,0].data.cpu().numpy())
}
utils.summarize(
writer=writer,
global_step=global_step,
images=image_dict,
scalars=scalar_dict)
if global_step % hps.train.eval_interval == 0:
evaluate(hps, net_g, eval_loader, writer_eval)
utils.save_checkpoint(net_g, None, hps.train.learning_rate, epoch, os.path.join(hps.model_dir, "G_{}.pth".format(global_step)))
utils.save_checkpoint(net_g, None, hps.train.learning_rate, epoch,
os.path.join(hps.model_dir, "G_latest.pth"))
# utils.save_checkpoint(net_d, optim_d, hps.train.learning_rate, epoch, os.path.join(hps.model_dir, "D_{}.pth".format(global_step)))
old_g=os.path.join(hps.model_dir, "G_{}.pth".format(global_step-4000))
# old_d=os.path.join(hps.model_dir, "D_{}.pth".format(global_step-400))
if os.path.exists(old_g):
os.remove(old_g)
# if os.path.exists(old_d):
# os.remove(old_d)
global_step += 1
if global_step == 4001:
exit()
if rank == 0:
logger.info('====> Epoch: {}'.format(epoch))
def evaluate(hps, generator, eval_loader, writer_eval):
generator.eval()
with torch.no_grad():
for batch_idx, (x, x_lengths, spec, spec_lengths, y, y_lengths, speakers) in enumerate(eval_loader):
x, x_lengths = x.cuda(0), x_lengths.cuda(0)
spec, spec_lengths = spec.cuda(0), spec_lengths.cuda(0)
y, y_lengths = y.cuda(0), y_lengths.cuda(0)
speakers = speakers.cuda(0)
# remove else
x = x[:1]
x_lengths = x_lengths[:1]
spec = spec[:1]
spec_lengths = spec_lengths[:1]
y = y[:1]
y_lengths = y_lengths[:1]
speakers = speakers[:1]
break
y_hat, attn, mask, *_ = generator.module.infer(x, x_lengths, speakers, max_len=1000)
y_hat_lengths = mask.sum([1,2]).long() * hps.data.hop_length
mel = spec_to_mel_torch(
spec,
hps.data.filter_length,
hps.data.n_mel_channels,
hps.data.sampling_rate,
hps.data.mel_fmin,
hps.data.mel_fmax)
y_hat_mel = mel_spectrogram_torch(
y_hat.squeeze(1).float(),
hps.data.filter_length,
hps.data.n_mel_channels,
hps.data.sampling_rate,
hps.data.hop_length,
hps.data.win_length,
hps.data.mel_fmin,
hps.data.mel_fmax
)
image_dict = {
"gen/mel": utils.plot_spectrogram_to_numpy(y_hat_mel[0].cpu().numpy())
}
audio_dict = {
"gen/audio": y_hat[0,:,:y_hat_lengths[0]]
}
if global_step == 0:
image_dict.update({"gt/mel": utils.plot_spectrogram_to_numpy(mel[0].cpu().numpy())})
audio_dict.update({"gt/audio": y[0,:,:y_lengths[0]]})
utils.summarize(
writer=writer_eval,
global_step=global_step,
images=image_dict,
audios=audio_dict,
audio_sampling_rate=hps.data.sampling_rate
)
generator.train()
if __name__ == "__main__":
main()
utils.py
+3 -3
View File
@@ -143,13 +143,13 @@ def load_filepaths_and_text(filename, split="|"):
def get_hparams(init=True):
parser = argparse.ArgumentParser()
parser.add_argument('-c', '--config', type=str, default="./configs/uma87.json",
parser.add_argument('-c', '--config', type=str, default="./configs/finetune_speaker.json",
help='JSON file for configuration')
parser.add_argument('-m', '--model', type=str, default="./pretrained_models/uma_G_0.pth",
parser.add_argument('-m', '--model', type=str, default="./OUTPUT_MODEL",
help='Model name')
args = parser.parse_args()
model_dir = os.path.join("../drive/MyDrive", args.model)
model_dir = args.model
if not os.path.exists(model_dir):
os.makedirs(model_dir)