Reading WAV file from TIMIT database in python
I am trying to read a wav file from TIMIT database in python, but I am getting an error:
When I use wave:
wave.Error: file does not start with RIFF id
When I use scipy:
ValueError: File format b'NIST'... not understood.
and when i use librosa the program gets stuck. I tried to convert it to wav using sox:
cmd = "sox " + wav_file + " -t wav " + new_wav
subprocess.call(cmd, shell=True)
and it didn't help. I saw an old answer referencing the scikits.audiolab package, but it looks like it is no longer supported.
How can I read this file to get an ndarray of data?
thank
source to share
Your file is not a WAV file. Apparently this is the NIST SPHERE file. From the LDC web page : "Many LDC corporations contain NIST SPHERE speech files." According to the NIST File Format , the first four characters of a NIST
. This is what scipy tells you: it doesn't know how to read a file that starts with NIST
.
I suspect that you need to convert the file to WAV if you want to read the file from any of the libraries you've tried.To force the conversion to WAV using a program sph2pipe
, use the command -f wav
(or, equivalently, -f rif
), for example
sph2pipe -f wav input.sph output.wav
source to share
issue this from command line to check its wav file ... or not
xxd -b myaudiofile.wav | head
if its wav format looks something like
00000000: 01010010 01001001 01000110 01000110 10111100 10101111 RIFF..
00000006: 00000001 00000000 01010111 01000001 01010110 01000101 ..WAVE
0000000c: 01100110 01101101 01110100 00100000 00010000 00000000 fmt ..
00000012: 00000000 00000000 00000001 00000000 00000001 00000000 ......
00000018: 01000000 00011111 00000000 00000000 01000000 00011111 @...@.
0000001e: 00000000 00000000 00000001 00000000 00001000 00000000 ......
00000024: 01100100 01100001 01110100 01100001 10011000 10101111 data..
0000002a: 00000001 00000000 10000001 10000000 10000001 10000000 ......
00000030: 10000001 10000000 10000001 10000000 10000001 10000000 ......
00000036: 10000001 10000000 10000001 10000000 10000001 10000000 ......
Please note that the wav file starts with RIFF characters which is a mandatory indicator, the file uses wav codec ... if your system (I'm on linux) does not have the above command line utility: xxd then use any hex editor like wxHexEditor to similarly examine your wav file to confirm that you see RIFF ... if there is no RIFF then its just not a wav file
Here are the details of the wav format specifications
http://soundfile.sapp.org/doc/WaveFormat/
http://www-mmsp.ece.mcgill.ca/Documents/AudioFormats/WAVE/WAVE.html
http://unusedino.de/ec64/technical/formats/wav.html
http://www.drdobbs.com/database/inside-the-riff-specification/184409308
http://www.topherlee.com/software/pcm-tut-wavformat.html
http://www.labbookpages.co.uk/audio/javaWavFiles.html
source to share