post
https://us-east-1.recall.ai/api/v1/bot//output_audio/
Recent Requests
Log in to see full request history
| Time | Status | User Agent | |
|---|---|---|---|
Retrieving recent requests… | |||
Loading…
This endpoint is meant to output small snippets of audio, such as alerts/tones and recording consent disclaimers.
Since it doesn't provide a high level of control and isn't optimized for low latency, you should not use it for conversational or interactive features involving bot audio. For these use cases you should instead use Output Media.
Use this endpoint to:
- Play short prerecorded audio clips into a live meeting, such as alerts, tones, or other brief sound cues.
- Deliver a brief recording consent disclaimer when the bot joins or when recording is about to begin.
- Play a very short spoken notice to let participants know the bot is present or a workflow step has started.
- Trigger simple audio prompts during a meeting, such as reminders, warnings, or next-step instructions.
What not to use this endpoint for:
- Do not use this endpoint for interactive agents that need to listen and respond during the call.
- Do not use this endpoint for dynamic speech generated in real time from an LLM, TTS system, or other runtime workflow.
- Do not use this endpoint for bots that need to provide context-aware answers based on live meeting audio or transcripts.
- Do not use this endpoint for continuous or low-latency conversational voice output that changes as the meeting unfolds.
- Do not use this endpoint when you need to stream audio or video from a webpage or app into the meeting in real time.