To activate the speech recognition on your IPCC server, you must use the following subsystem , a Media Resource Control Protocol (MRCP) Automated Speech Recognition (ASR) ( It is the client component).
If it comes with a separate ASR server like Nuance, you are able to enable speech recognition. So let’s review a little the conversation between the ASR server and IPCC server using MRCP.
MRCP is a mechanism which allows a client device (IPCC/phones) that requires audio stream processing to control processing resources such as ASR ad TTS servers ( for speech recognition and Text-to-speech conversion). Remind also that if you want to enable also Text-to-speech , you need to have another MRCP dedicated for TTS . So one MRCP ASR and one MRCP TTS.
MRCP relies on the Real Time Streaming protocol (RTSP) or SIP as a control protocol for setting up and controlling sessions. RSTP/SIP is also responsible for setting up media streams between the client and the server by using RTP (it is a kind of H245 negotiations)