ProDiff

ProDiff copied to clipboard

Model Training

4

Hi thanks for the repo, wanted to check in with the training metrics. Currently, i am training the ProDiff Teacher model and getting the following validation results. Saw that there...

keelezibel

Hi, thanks for your great work! I would like to ask if I want to apply the framework to a multi-speaker TTS task such as TTS on LibriTTS dataset, how...

ZhengRachel

I encountered a KeyError: 'pre_align_cls' while running the preprocess step with the command "python data_gen/tts/bin/pre_align.py". How do I fix it? thanks

1

phamdan

Missing base_pre_align.py

2

In egs/datasets/audio/libritts/prealign.py the importing "from data_gen.tts.base_pre_align import BasePreAlign". It seems that base_pre_align.py is missing in the repo. Could you please add this file? Thank you.

Hertin

Dear Rongjiehuang: Thanks for your teams opensource model. But your link of LJspeech seem doesn't work. Could u please reload data again? Thanks. Best,Regard. Shayne C.

CiE-XinYuChen

the link of your processed LJSpeech dataset is expired,could you send me the processed LJSpeech dataset?

fanmanqian

Dear Rongjie, I am a research assistant at the City University of Hong Kong, I currently conduct research related to neurolinguistics and appreciate your work about text to speech generation....

HLSUD

The original code causes an error, which is a bug.

beautifulrem

Update ProDiff.py

1

The key 'f0_std' is used in the code with the dictionary 'hparams', but this key cannot be found in the dictionary 'hparams', resulting in a KeyError error. The reason is...

beautifulrem

I'm confused about the objective assessment such as STOI and PESQ because they need the two input audios to have the same length. However, the predicted audio may have a...

zhenye234