You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Nice to see contrastive learning used in music area, is there any length limit? Is it possible to get meaningful representation (for example, hundreds dimension vector) of song (few minutes long) with this method? Look forward for your reply, thanks a lot.
The text was updated successfully, but these errors were encountered:
@Spijkervet Hello, sorry to disturb. I tried change the audio_length, and it failed. I read the source code and found that the encoder of audio is sample_cnn which can't process variable size input? How did you solve this? Looking forward to your reply, thanks a lot.
Nice to see contrastive learning used in music area, is there any length limit? Is it possible to get meaningful representation (for example, hundreds dimension vector) of song (few minutes long) with this method? Look forward for your reply, thanks a lot.
The text was updated successfully, but these errors were encountered: