Data set dimension issues during training

sp-uhh / sgmse

Score-based Generative Models (Diffusion Models) for Speech Enhancement and Dereverberation

MIT License

454 stars 69 forks source link

Data set dimension issues during training #43

Closed SevenXander closed 2 months ago

SevenXander commented 6 months ago

Expected 3D (unbatched) or 4D (batched) input to conv2d, but got input of size [3, 4, 256, 256, 2] 。how to solve？

julius-richter commented 2 months ago

This error might occur because you are trying to enhance stereo files with two audio channels. However, the method only works with mono audio (1 channel).