VITS HiFiTTS doc (NVIDIA#6288)

* Added VITS documentation * Typos * Added experimental note * Updated tutorial * Added spectrogram visualization * Updated ipa_cmudict version --------- Signed-off-by: Evgeny Shabalin <[email protected]> Co-authored-by: Xuesong Yang <[email protected]> Signed-off-by: hsiehjackson <[email protected]>
hsiehjackson · Jun 2, 2023 · fa333f0 · fa333f0
1 parent 2732e2f
commit fa333f0
Show file tree

Hide file tree

Showing 4 changed files with 482 additions and 3 deletions.
diff --git a/docs/source/tts/data/ngc_models_e2e.csv b/docs/source/tts/data/ngc_models_e2e.csv
@@ -1,2 +1,3 @@
 Locale,Model Name,Dataset,Sampling Rate,#Spk,Phoneme Unit,Model Class,Overview,Checkpoint
-en-US,tts_en_lj_vits,LJSpeech,22050Hz,1,IPA,nemo.collections.tts.models.vits.VitsModel,`tts_en_lj_vits <https://ngc.nvidia.com/catalog/models/nvidia:nemo:tts_en_lj_vits>`_,``https://api.ngc.nvidia.com/v2/models/nvidia/nemo/tts_en_lj_vits/versions/1.13.0/files/vits_ljspeech_fp16_full.nemo``
+en-US,tts_en_lj_vits,LJSpeech,22050Hz,1,IPA,nemo.collections.tts.models.vits.VitsModel,`tts_en_lj_vits <https://ngc.nvidia.com/catalog/models/nvidia:nemo:tts_en_lj_vits>`_,``https://api.ngc.nvidia.com/v2/models/nvidia/nemo/tts_en_lj_vits/versions/1.13.0/files/vits_ljspeech_fp16_full.nemo``
+en-US,tts_en_hifitts_vits,HiFiTTS,44100Hz,10,IPA,nemo.collections.tts.models.vits.VitsModel,`tts_en_hifitts_vits <https://ngc.nvidia.com/catalog/models/nvidia:nemo:tts_en_hifitts_vits>`_,``https://api.ngc.nvidia.com/v2/models/nvidia/nemo/tts_en_hifitts_vits/versions/r1.15.0/files/vits_en_hifitts.nemo``
diff --git a/examples/tts/vits.py b/examples/tts/vits.py
@@ -14,7 +14,6 @@
 
 import pytorch_lightning as pl
 
-from nemo.collections.common.callbacks import LogEpochTimeCallback
 from nemo.collections.tts.models.vits import VitsModel
 from nemo.core.config import hydra_runner
 from nemo.utils.exp_manager import exp_manager
@@ -26,7 +25,7 @@ def main(cfg):
     exp_manager(trainer, cfg.get("exp_manager", None))
     model = VitsModel(cfg=cfg.model, trainer=trainer)
 
-    trainer.callbacks.extend([pl.callbacks.LearningRateMonitor(), LogEpochTimeCallback()])
+    trainer.callbacks.extend([pl.callbacks.LearningRateMonitor()])
     trainer.fit(model)
 
 

diff --git a/nemo/collections/tts/models/vits.py b/nemo/collections/tts/models/vits.py
@@ -394,6 +394,13 @@ def list_available_models(cls) -> 'List[PretrainedModelInfo]':
             class_=cls,
         )
         list_of_models.append(model)
+        model = PretrainedModelInfo(
+            pretrained_model_name="tts_en_hifitts_vits",
+            location="https://api.ngc.nvidia.com/v2/models/nvidia/nemo/tts_en_hifitts_vits/versions/r1.15.0/files/vits_en_hifitts.nemo",
+            description="This model is trained on HiFITTS sampled at 44100Hz with and can be used to generate male and female English voices with an American accent.",
+            class_=cls,
+        )
+        list_of_models.append(model)
         return list_of_models
 
     @typecheck(