Re: Save incoming audio bytes from UniMRCP

[Ken Walker]

What is the integration point with UniMRCP? Is there some documentation on what the packets are? I.E. is it straight ulaw, or Linear16? or is it RTP?

I have done some code in the past for capturing both types and writing to a WAV file that I create the header for, but knowing the payload is key to making it work.

Ken

On Thursday, May 9, 2024 at 4:02:30 AM UTC-7 Sonu wrote:

Dear Unimrcp Community,

I've integrated Unimrcp with the socket API and can receive speech signals from users. However, when saving these speech byte signals as WAV files, the audio plays as a "ZAZZ" sound instead of the user's actual speech.

I've tried various approaches to save the byte data as a WAV file, such as constructing the header manually and combining it with the audio data, but the resulting files don't play the correct audio.

Could someone please provide guidance on how to properly save this byte data as a WAV file so that it plays the correct audio spoken by the user?

Any help or insights would be greatly appreciated. Thank you in advance!

[Vahagn Kocharyan]

Hi

it is configuration problem. Both side must have same configuration.

   <codecs own-preference="false">PCMU PCMA G722 L16/96/8000 telephone-event/101/8000</codecs>
      <!-- <codecs own-preference="false">PCMU PCMA G722 L16/96/8000 PCMU/97/16000 PCMA/98/16000 L16/99/16000 AMR-WB/102/16000/1/octet-align=1</codecs> -->
      <
see this.