Open dan-homebrew opened 1 month ago
should we use current speech enhancement tech to tackle this instead of developing our own?
We should very much use out of the box speech enhancement tech. The problem is general enough that the ichigo use-case is not differentiated enough to warrant domain-specific fine-tuning.
@PodsAreAllYouNeed we can icebox this if this is not priority
also cc @dan-homebrew
Speech Enhancement is also included in the VADHandler proposed in #91
Goal