Enabling whole call recording
Voice Platform records caller utterances as waveform logs—audio recordings of what the caller said at each point where the application prompted him or her for input. However, these waveforms are each separate files, and do not include the TTS or audio prompt that preceded the caller utterance. To store an audio recording of the entire call from start to finish in an additional file, you must enable whole call recording.
The whole call recording feature is governed by a group of service properties that may be set for the Speech Server via the Management Station.
- server.rtp.wcr.enable—Enables whole call recording.
- server.rtp.wcr.maxminutes—Specifies the maximum minutes allowed for one call.
- server.rtp.wcr.outputtype—Specifies whether the call is saved in a single file, or two files (one for the caller's audio, one for the application's audio).
- server.rtp.wcr.sampling—Specifies the percent chance that a given call is recorded.
