search

Plachtaa / VITS-fast-fine-tuning

This repo is a pipeline of VITS finetuning for fast speaker adaptation TTS, and many-to-many voice conversion
Apache License 2.0
4.65k stars 698 forks source link

step3.5出错 #567

Open youyi0218 opened 6 months ago

youyi0218 commented 6 months ago

AssertionError Traceback (most recent call last) /content/VITS-fast-fine-tuning/preprocess_v2.py in 34 if speaker not in speakers: 35 speakers.append(speaker) ---> 36 assert (len(speakers) != 0), "No audio file found. Please check your uploaded file structure." 37 # Source 3 (Optional): sampled audios as extra training helpers 38 if args.add_auxiliary_data:

AssertionError: No audio file found. Please check your uploaded file structure.

wodhei commented 5 months ago

解决了吗？我也出现一样的状况。

yubinyes commented 2 months ago

我单步调试，用的标贝数据，到了_, probs = model.detect_language(mel)这一行都异常了，最后的报错都是no short audios found。我看了一下，用的whisper large出错， whisper medium可以生成正常的short_character_anno.txt。不知道为什么large会出错。

fesolla commented 2 months ago

同样的状况+1