Pregunta

I am trying to integrate my telephony application with pocket sphinx speech recognizer. The recorded file type which I get is of the following type:

Input File     : '5555_9911848379__15-10-14-17-36_.wav'
Channels       : 1
Sample Rate    : 11025
Precision      : 8-bit
Duration       : 00:00:03.99 = 44000 samples ~ 299.32 CDDA sectors
File Size      : 44.0k
Bit Rate       : 88.3k
Sample Encoding: 8-bit Unsigned Integer PCM

but, the sample program which is given here http://cmusphinx.sourceforge.net/wiki/tutorialpocketsphinx expects 16 bit PCM audio sampled at 16000 HZ.

Is there any way I can decode the above mentioned recorded file. Or will I have to convert the above format to the 16 bit format?

¿Fue útil?

Solución

Is there any way I can decode the above mentioned recorded file.

No, you need 16-bit

Or will I have to convert the above format to the 16 bit format?

Conversion will not help since information is lost already. You need 16-bit recording.

You need to make sure you are recording audio in a proper format.

Licenciado bajo: CC-BY-SA con atribución
No afiliado a StackOverflow
scroll top