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

偶尔不能按模式三克隆原声 #279

Open
qunyinghuicui-copy opened this issue Nov 22, 2024 · 7 comments
Open

偶尔不能按模式三克隆原声 #279

qunyinghuicui-copy opened this issue Nov 22, 2024 · 7 comments
Labels
bug Something isn't working

Comments

@qunyinghuicui-copy
Copy link

qunyinghuicui-copy commented Nov 22, 2024

Attempt 1 failed, retrying...
TTS request failed, status code: 404
TTS request failed, switching back to mode 2 and retrying
TTS request failed, status code: 404
❌ Error: Failed to get audio duration: list index out of range
Attempt 2 failed, retrying...
TTS request failed, status code: 404
TTS request failed, switching back to mode 2 and retrying
TTS request failed, status code: 404
❌ Error: Failed to get audio duration: list index out of range
❌ Error in warmup: Failed to generate audio after 3 attempts: Generated audio duration is 0
🔄 Generating TTS audio... ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0% -:--:--
2024-11-22 21:48:59.690 Uncaught app exception
Traceback (most recent call last):
File "D:\dub\VideoLingo\core\all_tts_functions\tts_main.py", line 52, in tts_main
raise Exception("Generated audio duration is 0")
Exception: Generated audio duration is 0

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
File "D:\Anaconda\envs\videolingo\lib\site-packages\streamlit\runtime\scriptrunner\exec_code.py", line 88, in exec_func_with_error_handling
result = func()
File "D:\Anaconda\envs\videolingo\lib\site-packages\streamlit\runtime\scriptrunner\script_runner.py", line 590, in code_to_exec
exec(code, module.dict)
File "D:\dub\VideoLingo\st.py", line 123, in
main()
File "D:\dub\VideoLingo\st.py", line 120, in main
audio_processing_section()
File "D:\dub\VideoLingo\st.py", line 79, in audio_processing_section
process_audio()
File "D:\dub\VideoLingo\st.py", line 99, in process_audio
step10_gen_audio.gen_audio()
File "D:\dub\VideoLingo\core\step10_gen_audio.py", line 207, in gen_audio
tasks_df = generate_tts_audio(tasks_df)
File "D:\dub\VideoLingo\core\step10_gen_audio.py", line 99, in generate_tts_audio
raise e
File "D:\dub\VideoLingo\core\step10_gen_audio.py", line 94, in generate_tts_audio
number, real_dur = process_row(row, tasks_df)
File "D:\dub\VideoLingo\core\step10_gen_audio.py", line 78, in process_row
tts_main(line, temp_file, number, tasks_df)
File "D:\dub\VideoLingo\core\all_tts_functions\tts_main.py", line 56, in tts_main
raise Exception(f"Failed to generate audio after {max_retries} attempts: {str(e)}")
Exception: Failed to generate audio after 3 attempts: Generated audio duration is 0

@qunyinghuicui-copy qunyinghuicui-copy closed this as not planned Won't fix, can't repro, duplicate, stale Nov 22, 2024
@qunyinghuicui-copy qunyinghuicui-copy changed the title 找不到 GPT-SoVITS-v2 目录或配置文件路径 生成不了TTS Nov 22, 2024
@qunyinghuicui-copy
Copy link
Author

屏幕截图 2024-11-22 215429
屏幕截图 2024-11-22 215434

@Huanshere
Copy link
Owner

可以先使用 GPT-SoVits 自己的 webUI 看看电脑能否正常使用这个模型推理语音

@qunyinghuicui-copy
Copy link
Author

可以先使用 GPT-SoVits 自己的 webUI 看看电脑能否正常使用这个模型推理语音

能正常推理

@qunyinghuicui-copy
Copy link
Author

屏幕截图 2024-11-23 201915

@qunyinghuicui-copy qunyinghuicui-copy changed the title 生成不了TTS 偶尔不能按模式三克隆原声 Nov 25, 2024
@qunyinghuicui-copy
Copy link
Author

屏幕截图 2024-11-25 213623
能解决吗,有什么按模式二克隆的声音有点怪

@Huanshere
Copy link
Owner

这些具体的错误似乎和参考音频或者文本有关?我没有深入探讨过单独失败的原因,如果你有时间单独挑出失败的那个然后找到原因,欢迎贡献代码~

@Huanshere
Copy link
Owner

是的最近有不少用户反馈,原因是 gptsovits 对于 refer audio 有一定的要求,比如时长比如 prompt,这个需要详细的优化。。。

@Huanshere Huanshere added the bug Something isn't working label Dec 4, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
bug Something isn't working
Projects
None yet
Development

No branches or pull requests

2 participants