Open saikrishnarallabandi opened 5 years ago
@saikrishnarallabandi
@nishantgurunath feel free to add more details
@nishantgurunath cud confirm the issue.
Resynthesized file here: http://tts.speech.cs.cmu.edu/rsk/misc_stuff/B01___01_Matthew_____ADHBSUN2DA_00005_reconstructed.wav
Looks like this is due to the spectral representation we are using. Let me look at other representations
Extracting features at different frame shifts doesnt resolve the issue:
It seems to get corrupted on extraction and resynthesis Music becomes noise
ADHBSU/aligned/wav/B01_01_Matthew___ADHBSUN2DA_00005.wav