search

open-mmlab / Amphion

Amphion (/æmˈfaɪən/) is a toolkit for Audio, Music, and Speech Generation. Its purpose is to support reproducible research and help junior researchers and engineers get started in the field of audio, music, and speech generation research and development.
https://openhlt.github.io/amphion/
MIT License
4.45k stars 379 forks source link

[BUG]: prompt_examples/*.wav missing in NaturalSpeech2 #132

Closed xiangmy21 closed 7 months ago

xiangmy21 commented 7 months ago

Describe the bug

when use NaturalSpeech2 inference, it report RuntimeError: not found NaturalSpeech2/prompt_example/ref_audio.wav. same as https://github.com/open-mmlab/Amphion/issues/114, but in NaturalSpeech2, not VALLE. I think the problem comes from the .gitignore as well, it ignores all *.wav expect prompt_examples in VALLE now.

How To Reproduce

follow https://huggingface.co/amphion/naturalspeech2_libritts .

Expected behavior

such as:

RuntimeError: Failed to open the input "~/amphion/egs/tts/NaturalSpeech2/prompt_example/ref_audio.wav" (No such file or directory).
xiangmy21 commented 7 months ago

Oh I get it, not a problem. Just download from the demo page https://speechresearch.github.io/naturalspeech2/ or use own voice. But better provide a default example. Thanks~

lmxue commented 7 months ago

Thanks for your attention and support towards our work. I will close this issue. If you have any further questions about prompt_examples, please feel free to re-open this issue. We are glad to follow up!