Voice Verify is designed for the following voice inputs:
- SIP calls with voice transferred via RTP streams
- HTTP streams
- WebSocket streams
Voice Verify expects to receive customer’s voice only. This requires forking of the call stream and can be achieved via SIPREC or some other means.
Once the call is received, it is processed with Speaker Identification voice biometric technology. It can be either enrolled or verified against previously saved voiceprint. The results are provided via RESTful API.
Additional info and latest technical documentation including hardware requirements can be found here.