search

shansongliu / MU-LLaMA

MU-LLaMA: Music Understanding Large Language Model
GNU General Public License v3.0
221 stars 16 forks source link

mp3 audio format #17

Open wwfcnu opened 11 months ago

wwfcnu commented 11 months ago

First of all, thank you very much for your work. In addition, will you consider supporting mp3 format in the future? Use ffmpeg instead of torchaudio to support more formats

shansongliu commented 11 months ago

First of all, thank you very much for your work. In addition, will you consider supporting mp3 format in the future? Use ffmpeg instead of torchaudio to support more formats

We have just updated our gradio demo and inference scripts recently. I have tested mp3 file input for the gradio demo, it works well, you can have a try.

wwfcnu commented 11 months ago

First of all, thank you very much for your work. In addition, will you consider supporting mp3 format in the future? Use ffmpeg instead of torchaudio to support more formats

We have just updated our gradio demo and inference scripts recently. I have tested mp3 file input for the gradio demo, it works well, you can have a try.

image
shansongliu commented 10 months ago

First of all, thank you very much for your work. In addition, will you consider supporting mp3 format in the future? Use ffmpeg instead of torchaudio to support more formats

We have just updated our gradio demo and inference scripts recently. I have tested mp3 file input for the gradio demo, it works well, you can have a try.

image

We will support it later. Stay tuned.

wwfcnu commented 10 months ago

@shansongliu I see that utils.py resamples the audio to 24000hz. What is the basis for this?