Added torchaudio.models.Tacotron2()
#669
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
NiklasGustafsson
suggested changes
Aug 1, 2022
| RegisterComponents(); | ||
| } | ||
|
|
||
| public (Tensor, Tensor, Tensor, Tensor) forward( |
There was a problem hiding this comment.
Doc comments on all public methods, please.
| } | ||
| } | ||
|
|
||
| public class Prenet : nn.Module |
There was a problem hiding this comment.
A public class -- please add doc comments.
| /// Tacotron2 model based on the implementation from | ||
| /// Nvidia https://github.com/NVIDIA/DeepLearningExamples/. | ||
| /// </summary> | ||
| public class Tacotron2 : nn.Module |
There was a problem hiding this comment.
I added a couple of comments below, but there are public methods and classes in this class that should have doc comments.
|
@kaiidams -- I'm going to make another release this week, since someone found a very serious performance bug in tensor creation. I'd love to get this PR in the release, but no pressure -- if you don't have time to work on it, I'll include it without doc comments, and then create an issue that I'll assign to you to add them later. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Added Tacotron2 implementation from https://github.com/pytorch/audio/blob/e502df0106403f7666f89fee09715256ea2e0df3/torchaudio/models/tacotron2.py
The file has a LICENSE notice from NVIDIA.
PyTorch has four pretrained mode
Quality of Griffin-Lim is not as good as WaveRNN. Phoneme based requires DeepPhonemizer to process texts which is written with PyTorch.
One of pretrained models from PyTorch converted to TorchSharp.
tacotron2_english_phonemes_1500_epochs_wavernn_ljspeech.pthhttps://drive.google.com/file/d/11TnhmCSUy7aO1pv9CBi7Qivhz25nl_2l/view?usp=sharing