search

HarryVolek / PyTorch_Speaker_Verification

PyTorch implementation of "Generalized End-to-End Loss for Speaker Verification" by Wan, Li et al.
BSD 3-Clause "New" or "Revised" License
575 stars 165 forks source link

getting 0 shape array in save_spectrogram_tisv (vctk dataset ) #57

Closed pranshurastogi29 closed 4 years ago

pranshurastogi29 commented 4 years ago

hey , i am getting confused in save_spectrogram_tisv function because it gives 0 shape of array produced for each speaker in vctk while it works just fine when i use timit dataset (concentrate on shapes) please help needed asap

61297732 c023fc80 a7d4 11e9 8020 48f6a925d3fd PNG PNG Image 1075 × 764 pixels Scaled 98

pranshurastogi29 commented 4 years ago

got the answer use db=100 in librosa.effects.split for vctk else 30 would work fine for timit