I am building an application using call recordings from asterisk. My problem is that I’m doing speech recognition based off a model that was trained on a sampling rate of 44.1kHz. However, when I Monitor() my calls, the recordings are sampled at 8kHz. I tried sox but the audio quality is very poor. Is there a way to change the sampling rate in asterisk.
Thanks in advance.
Basically no. You need to train at 8kHz. There may be exceptions, but public telephone lines offer usually only 8kHz.
This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.