search

dtjchen / spoken-command-processor

Neural network-based speech transcription model. Built with Keras (Python) and trained with TIMIT.
19 stars 6 forks source link

Argument mix-up in call to wavfile_to_mfccs #7

Open skoppula opened 7 years ago

skoppula commented 7 years ago

Hi!

Firstly, thanks for releasing this work on GitHub -- it's been super useful in my research.

The problem seems to be in the call to wavfile_to_mfccs in the TIMIT pre-processing here

mfccs_and_deltas, segment_duration_frames, hop_duration_frames = utils.wavfile_to_mfccs(wavfile)

But the return values of wavfile_to_mfccs have the last two arguments switched (here):

return mfccs_and_deltas, hop_length, n_fft

If you want, I can put in a one-line PR fixing this.

ngragaei commented 6 years ago

Hey, I need help. Can you help me please? @skoppula