search

microsoft / CLAP

Learning audio concepts from natural language supervision
MIT License
455 stars 35 forks source link

batch embedding #12

Closed xuyaoxun closed 11 months ago

xuyaoxun commented 1 year ago

Hi, I'm having problems with clap for model training, I'm feeding one audio at a time and getting embedding, how can I feed a batch of audio at once and get a representation of the batch?

bmartin1 commented 12 months ago

Hi, did you try the functions that take _batchsize as a parameter inside src/CLAPWrapper.py (e.g. get_audio_embeddings_per_batch())?

bmartin1 commented 11 months ago

Closing due to inactivity.