I need to integrate voice coming from SIP or SIPREC session to Micorsoft Speech or MS Bot. According to https://docs.microsofttranslator.com/speech-translate.html the voice should be streamed single channel, signed 16bit PCM audio sampled at 16 kHz. So seems need also to "translate" those packets from whatever codec to PCM. What is reccomended approach?
Asked
Active
Viewed 111 times