Note
Access to this page requires authorization. You can try signing in or changing directories.
Access to this page requires authorization. You can try changing directories.
To support real-time scenarios, like Virtual Agent and Agent Assist in call centers, an integration with the call center's telephony system is required.
Typically, integration with the Speech service is handled by a telephony client connected to the customers SIP/RTP processor, for example, to a Session Border Controller (SBC).
Usually the telephony client handles the incoming audio stream from the SIP/RTP processor, the conversion to PCM and connects the streams using continuous recognition. It also triages the processing of the results. For example, analysis of speech transcripts for Agent Assist or connect with a dialog processing engine (for example, Azure Botframework or Power Virtual Agent) for Virtual Agent.
For easier integration the Speech service also supports "ALAW in WAV container" and "MULAW in WAV container" for audio streaming. To build this integration, we recommend using the Speech SDK.