anicolson/spn-asi

Sum-Product Networks (SPNs) for Robust Automatic Speaker Identification.

spn-asi's Issues

ValueError while processing clean speech data.

Hi, while running main.py for clean data I am getting below error.
Can you guys please suggest any work around, I am using TIMIT dataset which contains wav, phn, wrd and txt file. Is it some issue due to that?
/SPN-Spk-Rec$ sudo python3 main.py --ver '0a' --test_clean_speech 1
Creating speaker list, as no pickle file exists...
Traceback (most recent call last):
File "main.py", line 180, in
spn_args = add_args(spn_args)
File "main.py", line 161, in add_args
train_clean_speech, train_clean_speech_len, _, _ = spn_batch._batch(path, ['si*.wav', 'sx*.wav'] , []) # clean training waveforms and lengths.
File "/home/msds18014/SPN-Spk-Rec/spn_batch.py", line 43, in batch
(, wav) = read(fpath) # read waveform from given file path.
File "/usr/local/lib/python3.5/dist-packages/scipy/io/wavfile.py", line 236, in read
file_size, is_big_endian = _read_riff_chunk(fid)
File "/usr/local/lib/python3.5/dist-packages/scipy/io/wavfile.py", line 168, in _read_riff_chunk
"understood.".format(repr(str1)))
ValueError: File format b'NIST'... not understood.