Real-time transcription for a live Zoom Phone call - Scribe API source + RTMS for Zoom Phone roadmap?

Realtime Media Streams
,
1

We ingest Zoom Phone events and post-call recordings/transcripts today, and we want to add real-time, in-call features (live agent guidance / script-position tracking) for our Zoom Phone dialer + combiner floor. Our understanding:

  • Zoom Phone has on-screen “Call Live Transcription”, but no live-transcript API/webhook. Programmatic transcription is post-call only (recording.completed → transcription_url).
    • RTMS provides live audio + transcript for Meetings, Video SDK, and Contact Center, but not Zoom Phone.
  • Two questions:
  1. The new Scribe API (GET /v2/aiservices/scribe/live) - can it take a live Zoom Phone call as its source, or is it a bring-your-own-audio streaming STT engine? If BYO-audio, is there a supported way to obtain the live audio of an in-progress Zoom Phone call?
  2. Is RTMS support for Zoom Phone on the roadmap (live audio/transcript for a Phone call, as exists for Meetings/Video SDK/Contact Center)?

Goal: determine whether real-time in-call transcription for Zoom Phone is achievable natively, or whether it requires Zoom Contact Center. Thanks!