Communications·production

IAM.Voice

Voice agents, scenarios, campaigns and operator hand-off.

Landing Application
IAM.Voice interface
Current public IAM.Voice surface. The screenshot was verified when the documentation was published.

Key capabilities

  • Inbound and outbound voice scenarios
  • Streaming ASR/TTS, barge-in and scenario engine
  • Campaigns, telephony and operator hand-off

Role in the ecosystem

IAM.Voice manages incoming and outgoing voice scripts. It connects telephony IAM.Comm, streaming speech components, dialog logic and operator hand-off in one observed scenario.

Main scenario

  1. Configure the number, trunk and valid communication hours.
  2. Collect the script: greeting, intent, actions and transfer conditions.
  3. Select the ASR/TTS profile and check the delay on a test call.
  4. Launch an inbound flow or a coordinated outbound campaign.
  5. Analyze the outcome, reasons for termination and quality of the dialogue.

Real-time circuit

VAD determines speech boundaries, ASR generates a streaming hypothesis, the script selects next action, and TTS returns audio. Barge-in interrupts synthesis when speaking user. The timeout of each stage is limited so that the AI ​​provider does not fail kept the call unanswered.

Data and security

Recording, transcription and personal data are included in accordance with the policy and consent. Contact lists are not mixed between campaigns/tenants. The operator receives the transfer context without provider credentials and hidden service instructions.

Analytics

For in-depth transcription, diarization and QA, recordings are transferred to IAM.Speech with the same correlation id.