Available samping rate 8000 is too far from requested 14000

Speech Recognition Toolkit

Brought to you by: air, arthchan2003, awb, bhiksha, and 5 others

This project can now be found here.

Available samping rate 8000 is too far from requested 14000

Forum: Help

Creator: Christian

Created: 2016-06-20

Updated: 2016-06-22

Christian - 2016-06-20

Hello,
I tried to use pocketsphinx with this command :

pocketsphinx_continuous -dict /usr/local/share/pocketsphinx/model/lm/fr_FR/frenchWords62K.dic -hmm /usr/local/share/pocketsphinx/model/hmm/fr_FR/french_f0 -lm /usr/local/share/pocketsphinx/model/lm/fr_FR/french-small.lm -adcdev hw:1,0 -inmic yes

Without success :( I have this error :

Available samping rate 8000 is too far from requested 14000

Somebody can help me please ?
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
- Nickolay V. Shmyrev - 2016-06-20
  
  Your hardware does not support recording at 16khz, model requires that rate. You need to make sure your driver is configured properly, for example, you might want to make sure your alsa is configured with dmix plugin. Or try pulseaudio instead of alsa.
  
  If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
  - Nickolay V. Shmyrev - 2016-06-20
    
    Or you can train 8khz model from Voxforge data.
    
    If you would like to refer to this comment somewhere else in this project, copy and paste the following link:

Christian - 2016-06-22

Thanks for your replies.

"pulseAudio -D" it's ok, but I think I will buy another micro because the results is not so good.

If you would like to refer to this comment somewhere else in this project, copy and paste the following link:

Log in to post a comment.