Asterisk + Kaldi Realtime Speech Recognition
Budget: $15 – $25 USD
Hi everyone,
I'm looking for experienced VoIP Engineer to give us helping hand on real-time speech recognition system we are building.
Our stack: Debian, Asterisk v.16-19, Kaldi + Vosk, Python/Javascript (node.js)
The main problem is that we are unable to pick a real-time stream of RTP which is a plain UDP, transform it to Websocket data and send to the Kaldi server to recognize. We would prefer to get it done without Kamailio and RTPengine for now, just plain Asterisk possibilities like UnicastRTP, ARI etc.
So we are seeking for experienced Asterisk engineer who can give us a valuable hint, share experience and/or write some code for us. Thank you in advance.
I'm looking for experienced VoIP Engineer to give us helping hand on real-time speech recognition system we are building.
Our stack: Debian, Asterisk v.16-19, Kaldi + Vosk, Python/Javascript (node.js)
The main problem is that we are unable to pick a real-time stream of RTP which is a plain UDP, transform it to Websocket data and send to the Kaldi server to recognize. We would prefer to get it done without Kamailio and RTPengine for now, just plain Asterisk possibilities like UnicastRTP, ARI etc.
So we are seeking for experienced Asterisk engineer who can give us a valuable hint, share experience and/or write some code for us. Thank you in advance.