hifigan_for_sherpa

.gitattributes

@@ -33,3 +33,18 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+hifigan_for_sherpa/khadijah_LJ_v2_vocoder.wav filter=lfs diff=lfs merge=lfs -text
+hifigan_for_sherpa/khadijah_vctk_v2_vocoder.wav filter=lfs diff=lfs merge=lfs -text
+hifigan_for_sherpa/musa_LJ_v2_vocoder.wav filter=lfs diff=lfs merge=lfs -text
+hifigan_for_sherpa/musa_vctk_v2_vocoder.wav filter=lfs diff=lfs merge=lfs -text
+hifigan_for_sherpa/pretrained/LJ_FT_T2_V1/generator_v1 filter=lfs diff=lfs merge=lfs -text
+hifigan_for_sherpa/pretrained/LJ_FT_T2_V2/generator_v2 filter=lfs diff=lfs merge=lfs -text
+hifigan_for_sherpa/pretrained/LJ_FT_T2_V3/generator_v3 filter=lfs diff=lfs merge=lfs -text
+hifigan_for_sherpa/pretrained/LJ_V1/generator_v1 filter=lfs diff=lfs merge=lfs -text
+hifigan_for_sherpa/pretrained/LJ_V2/generator_v2 filter=lfs diff=lfs merge=lfs -text
+hifigan_for_sherpa/pretrained/LJ_V3/generator_v3 filter=lfs diff=lfs merge=lfs -text
+hifigan_for_sherpa/pretrained/UNIVERSAL_V1/do_02500000 filter=lfs diff=lfs merge=lfs -text
+hifigan_for_sherpa/pretrained/UNIVERSAL_V1/g_02500000 filter=lfs diff=lfs merge=lfs -text
+hifigan_for_sherpa/pretrained/VCTK_V1/generator_v1 filter=lfs diff=lfs merge=lfs -text
+hifigan_for_sherpa/pretrained/VCTK_V2/generator_v2 filter=lfs diff=lfs merge=lfs -text
+hifigan_for_sherpa/pretrained/VCTK_V3/generator_v3 filter=lfs diff=lfs merge=lfs -text

hifigan_for_sherpa/.gitattributes

@@ -0,0 +1,35 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text

hifigan_for_sherpa/README.md

@@ -0,0 +1,20 @@
+---
+license: cc0-1.0
+---
+
+# بسم الله
+
+ATTENTION: HIFIGAN IS RETIRED. USE VOCOS INSTEAD.
+
+1) Converted universal v1, v2, v3 22050Hz hifigan models from [here](https://github.com/jik876/hifi-gan) and [here](https://drive.google.com/drive/folders/1-eEYTB5Av9jNql0WGBlRoi-WH2J7bp5Y)
+
+2) Converted to onnx using [this](https://github.com/k2-fsa/icefall/blob/master/egs/ljspeech/TTS/matcha/export_onnx_hifigan.py)
+
+3) Added metadata as such using the above python script:
+{'model_type': 'hifigan', 'model_filename': 'hifigan_universal_22050hz_v1', 'sample_rate': 22050, 'version': 1, 'model_author': 'jik876', 'maintainer': 'k2-fsa', 'dataset': 'Universal', 'url1': 'https://github.com/jik876/hifi-gan', 'url2': ''}
+
+Note1: hifigan v2 is much smaller, faster and with lower memory need
+
+Note2: As LJ speech is a female single speaker dataset, hifigan based on it does not work well on men's voice. vctk models work well for both genders.
+
+Conclusion: Use vctk_v2!

hifigan_for_sherpa/hifigan_universal_22050hz_v1.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61dcee7065555aadd136dab133a8a9182a490706f52721cd7acfbc74871e88c5
+size 55750084

hifigan_for_sherpa/hifigan_vctk_22050hz_v1.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7479d71687759848ca3d3bc277fc73dd6cc0bec5f014929ae5982999fa0a48e1
+size 55750074

hifigan_for_sherpa/hifigan_vctk_22050hz_v2.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6a6240d23ee9f6c76e060aed1351d185c7e189cc8c8ffa4f9382ddd44a18b779
+size 3749664

hifigan_for_sherpa/hifigan_vctk_22050hz_v3.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b4e651fe2bbb3858d36d193a511b9977db164f6451bf66abae40c96bdfb073af
+size 5863679

hifigan_for_sherpa/khadijah_LJ_v2_vocoder.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7232b5dcbdc9d11d0077b0504a85410358e6275786678ff3d43a9bb53ab428cf
+size 164988

hifigan_for_sherpa/khadijah_vctk_v2_vocoder.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:011363f5f86c91974940574a3328cdd816cd2423ca389f04eef4d59d30ad5a03
+size 165690

hifigan_for_sherpa/musa_LJ_v2_vocoder.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:588dbd69c7cb1d9b79243d3a38560b468685a776ea549936ab91af383b24176f
+size 166748

hifigan_for_sherpa/musa_vctk_v2_vocoder.wav

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:059774656edaede3a5affc2ae94a3697627fca3d78d2153576c8d113e80a09a6
+size 167196

hifigan_for_sherpa/pretrained/LJ_FT_T2_V1/config.json

@@ -0,0 +1,38 @@
+{
+    "resblock": "1",
+    "num_gpus": 0,
+    "batch_size": 16,
+    "learning_rate": 0.0004,
+    "adam_b1": 0.8,
+    "adam_b2": 0.99,
+    "lr_decay": 0.999,
+    "seed": 1234,
+
+    "upsample_rates": [8,8,2,2],
+    "upsample_kernel_sizes": [16,16,4,4],
+    "upsample_initial_channel": 512,
+    "resblock_kernel_sizes": [3,7,11],
+    "resblock_dilation_sizes": [[1,3,5], [1,3,5], [1,3,5]],
+    "resblock_initial_channel": 256,
+
+    "segment_size": 8192,
+    "num_mels": 80,
+    "num_freq": 1025,
+    "n_fft": 1024,
+    "hop_size": 256,
+    "win_size": 1024,
+
+    "sampling_rate": 22050,
+
+    "fmin": 0,
+    "fmax": 8000,
+    "fmax_loss": null,
+
+    "num_workers": 4,
+
+    "dist_config": {
+        "dist_backend": "nccl",
+        "dist_url": "tcp://localhost:54321",
+        "world_size": 1
+    }
+}

hifigan_for_sherpa/pretrained/LJ_FT_T2_V1/generator_v1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:64aa99598e561596c69cb86f738890a8400fac97d367159ba6ee1bbb9e348cde
+size 55788858

hifigan_for_sherpa/pretrained/LJ_FT_T2_V2/config.json

@@ -0,0 +1,38 @@
+{
+    "resblock": "1",
+    "num_gpus": 0,
+    "batch_size": 16,
+    "learning_rate": 0.0004,
+    "adam_b1": 0.8,
+    "adam_b2": 0.99,
+    "lr_decay": 0.999,
+    "seed": 1234,
+
+    "upsample_rates": [8,8,2,2],
+    "upsample_kernel_sizes": [16,16,4,4],
+    "upsample_initial_channel": 128,
+    "resblock_kernel_sizes": [3,7,11],
+    "resblock_dilation_sizes": [[1,3,5], [1,3,5], [1,3,5]],
+    "resblock_initial_channel": 64,
+
+    "segment_size": 8192,
+    "num_mels": 80,
+    "num_freq": 1025,
+    "n_fft": 1024,
+    "hop_size": 256,
+    "win_size": 1024,
+
+    "sampling_rate": 22050,
+
+    "fmin": 0,
+    "fmax": 8000,
+    "fmax_loss": null,
+
+    "num_workers": 4,
+
+    "dist_config": {
+        "dist_backend": "nccl",
+        "dist_url": "tcp://localhost:54321",
+        "world_size": 1
+    }
+}

hifigan_for_sherpa/pretrained/LJ_FT_T2_V2/generator_v2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1d3aff09cc4cc060a804b8efea4093315c331eec5f81e8df094989c49a1fbec7
+size 3758167

hifigan_for_sherpa/pretrained/LJ_FT_T2_V3/config.json

@@ -0,0 +1,38 @@
+{
+    "resblock": "2",
+    "num_gpus": 0,
+    "batch_size": 16,
+    "learning_rate": 0.0004,
+    "adam_b1": 0.8,
+    "adam_b2": 0.99,
+    "lr_decay": 0.999,
+    "seed": 1234,
+
+    "upsample_rates": [8,8,4],
+    "upsample_kernel_sizes": [16,16,8],
+    "upsample_initial_channel": 256,
+    "resblock_kernel_sizes": [3,5,7],
+    "resblock_dilation_sizes": [[1,2], [2,6], [3,12]],
+    "resblock_initial_channel": 128,
+
+    "segment_size": 8192,
+    "num_mels": 80,
+    "num_freq": 1025,
+    "n_fft": 1024,
+    "hop_size": 256,
+    "win_size": 1024,
+
+    "sampling_rate": 22050,
+
+    "fmin": 0,
+    "fmax": 8000,
+    "fmax_loss": null,
+
+    "num_workers": 4,
+
+    "dist_config": {
+        "dist_backend": "nccl",
+        "dist_url": "tcp://localhost:54321",
+        "world_size": 1
+    }
+}

hifigan_for_sherpa/pretrained/LJ_FT_T2_V3/generator_v3

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dd94d0dc55ae761dde16e7f304be04435ba5f96ccadecff433ba5531da4bc319
+size 5870081

hifigan_for_sherpa/pretrained/LJ_V1/config.json

@@ -0,0 +1,38 @@
+{
+    "resblock": "1",
+    "num_gpus": 0,
+    "batch_size": 16,
+    "learning_rate": 0.0002,
+    "adam_b1": 0.8,
+    "adam_b2": 0.99,
+    "lr_decay": 0.999,
+    "seed": 1234,
+
+    "upsample_rates": [8,8,2,2],
+    "upsample_kernel_sizes": [16,16,4,4],
+    "upsample_initial_channel": 512,
+    "resblock_kernel_sizes": [3,7,11],
+    "resblock_dilation_sizes": [[1,3,5], [1,3,5], [1,3,5]],
+    "resblock_initial_channel": 256,
+
+    "segment_size": 8192,
+    "num_mels": 80,
+    "num_freq": 1025,
+    "n_fft": 1024,
+    "hop_size": 256,
+    "win_size": 1024,
+
+    "sampling_rate": 22050,
+
+    "fmin": 0,
+    "fmax": 8000,
+    "fmax_loss": null,
+
+    "num_workers": 4,
+
+    "dist_config": {
+        "dist_backend": "nccl",
+        "dist_url": "tcp://localhost:54321",
+        "world_size": 1
+    }
+}

hifigan_for_sherpa/pretrained/LJ_V1/generator_v1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bb4b0cb7f9df59b8e57bb2e51a1bede57b43e9f0454863e3971c491f255505e4
+size 55788858

hifigan_for_sherpa/pretrained/LJ_V2/config.json

@@ -0,0 +1,38 @@
+{
+    "resblock": "1",
+    "num_gpus": 0,
+    "batch_size": 16,
+    "learning_rate": 0.0002,
+    "adam_b1": 0.8,
+    "adam_b2": 0.99,
+    "lr_decay": 0.999,
+    "seed": 1234,
+
+    "upsample_rates": [8,8,2,2],
+    "upsample_kernel_sizes": [16,16,4,4],
+    "upsample_initial_channel": 128,
+    "resblock_kernel_sizes": [3,7,11],
+    "resblock_dilation_sizes": [[1,3,5], [1,3,5], [1,3,5]],
+    "resblock_initial_channel": 64,
+
+    "segment_size": 8192,
+    "num_mels": 80,
+    "num_freq": 1025,
+    "n_fft": 1024,
+    "hop_size": 256,
+    "win_size": 1024,
+
+    "sampling_rate": 22050,
+
+    "fmin": 0,
+    "fmax": 8000,
+    "fmax_loss": null,
+
+    "num_workers": 4,
+
+    "dist_config": {
+        "dist_backend": "nccl",
+        "dist_url": "tcp://localhost:54321",
+        "world_size": 1
+    }
+}

hifigan_for_sherpa/pretrained/LJ_V2/generator_v2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3fac378c5918fb2c102733f21eeaa8e9a4ca6cda24dbfddc55bbb947c78d562f
+size 3758167

hifigan_for_sherpa/pretrained/LJ_V3/config.json

@@ -0,0 +1,38 @@
+{
+    "resblock": "2",
+    "num_gpus": 0,
+    "batch_size": 16,
+    "learning_rate": 0.0002,
+    "adam_b1": 0.8,
+    "adam_b2": 0.99,
+    "lr_decay": 0.999,
+    "seed": 1234,
+
+    "upsample_rates": [8,8,4],
+    "upsample_kernel_sizes": [16,16,8],
+    "upsample_initial_channel": 256,
+    "resblock_kernel_sizes": [3,5,7],
+    "resblock_dilation_sizes": [[1,2], [2,6], [3,12]],
+    "resblock_initial_channel": 128,
+
+    "segment_size": 8192,
+    "num_mels": 80,
+    "num_freq": 1025,
+    "n_fft": 1024,
+    "hop_size": 256,
+    "win_size": 1024,
+
+    "sampling_rate": 22050,
+
+    "fmin": 0,
+    "fmax": 8000,
+    "fmax_loss": null,
+
+    "num_workers": 4,
+
+    "dist_config": {
+        "dist_backend": "nccl",
+        "dist_url": "tcp://localhost:54321",
+        "world_size": 1
+    }
+}

hifigan_for_sherpa/pretrained/LJ_V3/generator_v3

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b5e89fc0c45924525b7bd0c974aaf8b55aa8e0f9115a83356632d5aa11b8a554
+size 5870081

hifigan_for_sherpa/pretrained/UNIVERSAL_V1/config.json

@@ -0,0 +1,37 @@
+{
+    "resblock": "1",
+    "num_gpus": 0,
+    "batch_size": 16,
+    "learning_rate": 0.0002,
+    "adam_b1": 0.8,
+    "adam_b2": 0.99,
+    "lr_decay": 0.999,
+    "seed": 1234,
+
+    "upsample_rates": [8,8,2,2],
+    "upsample_kernel_sizes": [16,16,4,4],
+    "upsample_initial_channel": 512,
+    "resblock_kernel_sizes": [3,7,11],
+    "resblock_dilation_sizes": [[1,3,5], [1,3,5], [1,3,5]],
+
+    "segment_size": 8192,
+    "num_mels": 80,
+    "num_freq": 1025,
+    "n_fft": 1024,
+    "hop_size": 256,
+    "win_size": 1024,
+
+    "sampling_rate": 22050,
+
+    "fmin": 0,
+    "fmax": 8000,
+    "fmax_for_loss": null,
+
+    "num_workers": 4,
+
+    "dist_config": {
+        "dist_backend": "nccl",
+        "dist_url": "tcp://localhost:54321",
+        "world_size": 1
+    }
+}

hifigan_for_sherpa/pretrained/UNIVERSAL_V1/do_02500000

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9418cde901d2faed35376f1841e07ddbad6dc760abb4292d795e02df05f7b5cd
+size 960411640

hifigan_for_sherpa/pretrained/UNIVERSAL_V1/g_02500000

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:771eaf4876485a35e25577563d390c262e23c2421e4a8c929eacfde34a5b7a60
+size 55788858

hifigan_for_sherpa/pretrained/VCTK_V1/config.json

@@ -0,0 +1,38 @@
+{
+    "resblock": "1",
+    "num_gpus": 0,
+    "batch_size": 16,
+    "learning_rate": 0.0002,
+    "adam_b1": 0.8,
+    "adam_b2": 0.99,
+    "lr_decay": 0.999,
+    "seed": 1234,
+
+    "upsample_rates": [8,8,2,2],
+    "upsample_kernel_sizes": [16,16,4,4],
+    "upsample_initial_channel": 512,
+    "resblock_kernel_sizes": [3,7,11],
+    "resblock_dilation_sizes": [[1,3,5], [1,3,5], [1,3,5]],
+    "resblock_initial_channel": 256,
+
+    "segment_size": 8192,
+    "num_mels": 80,
+    "num_freq": 1025,
+    "n_fft": 1024,
+    "hop_size": 256,
+    "win_size": 1024,
+
+    "sampling_rate": 22050,
+
+    "fmin": 0,
+    "fmax": 8000,
+    "fmax_loss": null,
+
+    "num_workers": 4,
+
+    "dist_config": {
+        "dist_backend": "nccl",
+        "dist_url": "tcp://localhost:54321",
+        "world_size": 1
+    }
+}

hifigan_for_sherpa/pretrained/VCTK_V1/generator_v1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:27cdeb835874516f9404d6c1a9ea229b092fd98c329b8444f5955c24cf7b29a1
+size 55788858

hifigan_for_sherpa/pretrained/VCTK_V2/config.json

@@ -0,0 +1,38 @@
+{
+    "resblock": "1",
+    "num_gpus": 0,
+    "batch_size": 16,
+    "learning_rate": 0.0002,
+    "adam_b1": 0.8,
+    "adam_b2": 0.99,
+    "lr_decay": 0.999,
+    "seed": 1234,
+
+    "upsample_rates": [8,8,2,2],
+    "upsample_kernel_sizes": [16,16,4,4],
+    "upsample_initial_channel": 128,
+    "resblock_kernel_sizes": [3,7,11],
+    "resblock_dilation_sizes": [[1,3,5], [1,3,5], [1,3,5]],
+    "resblock_initial_channel": 64,
+
+    "segment_size": 8192,
+    "num_mels": 80,
+    "num_freq": 1025,
+    "n_fft": 1024,
+    "hop_size": 256,
+    "win_size": 1024,
+
+    "sampling_rate": 22050,
+
+    "fmin": 0,
+    "fmax": 8000,
+    "fmax_loss": null,
+
+    "num_workers": 4,
+
+    "dist_config": {
+        "dist_backend": "nccl",
+        "dist_url": "tcp://localhost:54321",
+        "world_size": 1
+    }
+}

hifigan_for_sherpa/pretrained/VCTK_V2/generator_v2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:049ff085eeae70ff0e1add39cfd4caad0fcca4511d259a84f68a0d1c1bbb2659
+size 3758167

hifigan_for_sherpa/pretrained/VCTK_V3/config.json

@@ -0,0 +1,38 @@
+{
+    "resblock": "2",
+    "num_gpus": 0,
+    "batch_size": 16,
+    "learning_rate": 0.0002,
+    "adam_b1": 0.8,
+    "adam_b2": 0.99,
+    "lr_decay": 0.999,
+    "seed": 1234,
+
+    "upsample_rates": [8,8,4],
+    "upsample_kernel_sizes": [16,16,8],
+    "upsample_initial_channel": 256,
+    "resblock_kernel_sizes": [3,5,7],
+    "resblock_dilation_sizes": [[1,2], [2,6], [3,12]],
+    "resblock_initial_channel": 128,
+
+    "segment_size": 8192,
+    "num_mels": 80,
+    "num_freq": 1025,
+    "n_fft": 1024,
+    "hop_size": 256,
+    "win_size": 1024,
+
+    "sampling_rate": 22050,
+
+    "fmin": 0,
+    "fmax": 8000,
+    "fmax_loss": null,
+
+    "num_workers": 4,
+
+    "dist_config": {
+        "dist_backend": "nccl",
+        "dist_url": "tcp://localhost:54322",
+        "world_size": 1
+    }
+}

hifigan_for_sherpa/pretrained/VCTK_V3/generator_v3

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ac38c592b20783ee5447269920ef8c79c13c79265258a4fecf89ef978cdd7caa
+size 5870081

hifigan_for_sherpa/source.txt

@@ -0,0 +1 @@
+https://huggingface.co/mah92/hifigan_for_sherpa