search

yh1008 / speech-to-text

mixlingual speech recognition system; hybrid (GMM+NNet) model; Kaldi + Keras
http://llcao.net/cu-deeplearning17/project.html
70 stars 19 forks source link

test data fix_data_dir.sh: no utterances remained: not proceeding further. #8

Closed yh1008 closed 7 years ago

yh1008 commented 7 years ago

utils/fix_data_dir.sh: filtered data/test/segments from 3652 to 0 lines based on filter /tmp/kaldi.i7EU/recordings. fix_data_dir.sh: no utterances remained: not proceeding further.

????? why my carefully constructed test set gets all filtered out ?????

yh1008 commented 7 years ago

cause the number of recording id in my wav.scp and segments are not the same... caused by a bug in my acoustic_data_prep.py script. Fixed now.