[asterisk-users] Speech To Text on linux with asterisk

Nickolay V. Shmyrev nshmyrev at nexiwave.com
Wed Sep 15 07:10:06 CDT 2010


2010/9/15, DHAVAL INDRODIYA <dhaval.it01034 at gmail.com>:
> Hello i have tried to convert through sphinx as suggested by Nickolay
>
> i am not getting convert my simple audio file.
>
> i am having following error while i fire following command
>
> pocketsphinx_continuous -infile /usr/etc/ask-propertyid.WAV -samprate 8000 \
> -hmm /usr/etcSpeechToText/Communicator_semi_40.cd_semi_6000 -lm
> lm_giga_20k_nvp_3gram.lm.DMP
>
> *FATAL_ERROR: "continuous.c", line 149: Failed to calibrate voice activity
> detection*

Hi

That's a progress already. I suspect this file has wrong format. It
must be little-endian 16-bit PCM with sample rate 8kHz. uLaw will not
work. It's also nice to have a little period of silence in the
beginning of the file.

Can you provide the file itself/share it somehow so I could take a look.



More information about the asterisk-users mailing list