Plachtaa / seed-vc

State-of-the-Art zero-shot voice conversion & singing voice conversion with in context learning
GNU General Public License v3.0
679 stars 76 forks source link

fix: last chunk crossfade dimension dismatch(#46) #54

Closed echonoshy closed 5 days ago

echonoshy commented 6 days ago

This error occurred during inference when the size of the last chunk's chunk2 was smaller than the overlap, causing a size mismatch during the addition operation.

I use source audio: 1:02s and reference audio: 0:09s,other parameters as default.

Error msg: ValueError: operands could not be broadcast together with shapes (5376,) (8192,)