Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[TTS]add vits network scripts, test=tts #1855

Merged
merged 5 commits into from May 24, 2022

Conversation

yt605155624
Copy link
Collaborator

@yt605155624 yt605155624 commented May 6, 2022

fix #1699

@yt605155624 yt605155624 marked this pull request as draft May 6, 2022 08:11
@yt605155624 yt605155624 added this to the r1.1.0 milestone May 6, 2022
@mergify
Copy link

mergify bot commented May 20, 2022

This pull request is now in conflict :(

@mergify mergify bot removed the conflicts label May 20, 2022
self.pad_mode = "reflect"

def _stft(self, wav: np.ndarray):
D = librosa.core.stft(
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

这里变量用D感觉是不是不太直观?

Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

这里 librosa stft 本身返回就是这么定义的

D : np.ndarray [shape=(1 + n_fft/2, n_frames), dtype=dtype]
Complex-valued matrix of short-term Fourier transform
coefficients.

paddlespeech/t2s/datasets/get_feats.py Show resolved Hide resolved
paddlespeech/t2s/models/vits/flow.py Outdated Show resolved Hide resolved
paddlespeech/t2s/models/vits/flow.py Outdated Show resolved Hide resolved
dropout_rate=flow_dropout_rate,
use_weight_norm=use_weight_norm_in_flow,
use_only_mean=use_only_mean_in_flow, )
# TODO: Add deterministic version as an option
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

这个是后续要加的还是已经加了忘记删掉注释?

Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

后续要加的,现在的 duration_predictor 是不确定性的,就是同样的输入每次的输出都不一样(VITS 的风格),类似于 FastSpeech2 的 duration_predictor 是确定的,同样的输入每次的结果都一样

paddlespeech/t2s/models/vits/flow.py Outdated Show resolved Hide resolved
paddlespeech/t2s/models/vits/residual_coupling.py Outdated Show resolved Hide resolved
@yt605155624 yt605155624 marked this pull request as ready for review May 24, 2022 06:21
@yt605155624 yt605155624 merged commit 5ee3cc0 into PaddlePaddle:develop May 24, 2022
@yt605155624 yt605155624 deleted the add_vits branch September 8, 2022 11:35
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
Status: Done
Development

Successfully merging this pull request may close these issues.

[TTS]VITS
2 participants