search

shansongliu / MU-LLaMA

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

Embedding of audio? #26

Open turian opened 3 months ago

turian commented 3 months ago

How can I use your model to get an embedding JUST of the audio file?

shansongliu commented 3 months ago

If you mean music embedding, you can just use the standalone MERT model, there is no need to use the whole MU-LLaMA model. If you want to use exactly the embedding that generated from our pipeline, then you may modify the model part to let it output the embedding by yourself.