***

id: 6826d07f-d320-45da-bff2-05230e1162a6
title: live\_transcribe
slug: /reference/live-transcribe
description: Transcribe a voice interaction in real-time.
max-toc-depth: 3
----------------

Start live transcription of the call. The transcription will be sent to the specified webhook URL.

## **Properties**

<ParamField path="live_transcribe" type="object" required={true} toc={true}>
  An object that accepts the following properties.
</ParamField>

<Indent>
  <ParamField path="live_transcribe.action" type="string | object" required={true} toc={true}>
    The action to perform. See [actions](#actions) below.
  </ParamField>
</Indent>

## **Actions**

The `action` property controls the transcription session. Use `start` to begin transcribing with configuration options, `stop` to end an active session, or `summarize` to request an on-demand AI summary mid-session.

<Tabs>
  <Tab title="start">
    <ParamField path="action.start" type="object" required={true} toc={true}>
      Start a live transcription session.
    </ParamField>

    <Indent>
      <ParamField path="start.webhook" type="string" toc={true}>
        The URL to receive transcription events via HTTP POST.
        When `live_events` is enabled, partial results are sent as they occur.
        When `ai_summary` is enabled, a summary is sent when the session ends.
        Authentication can also be set in the URL in the format of `username:password@url`.
      </ParamField>

      <ParamField path="start.lang" type="string" required={true} toc={true}>
        The language to transcribe.
        See [supported voices & languages](/docs/platform/voice/tts).
      </ParamField>

      <ParamField path="start.live_events" type="boolean" default="false" toc={true}>
        Whether to enable live events.
      </ParamField>

      <ParamField path="start.ai_summary" type="boolean" default="false" toc={true}>
        Whether to enable automatic AI summarization.
        When enabled, an AI-generated summary of the conversation will be sent to your webhook when the transcription session ends.
      </ParamField>

      <ParamField path="start.speech_timeout" type="integer" default="60000" toc={true}>
        The timeout for speech recognition in milliseconds. Minimum value: `1500`.
      </ParamField>

      <ParamField path="start.vad_silence_ms" type="integer" default="300 | 500" toc={true}>
        Voice activity detection silence time in milliseconds.
        Default depends on the speech engine: `300` for Deepgram, `500` for Google. Minimum value: `1`.
      </ParamField>

      <ParamField path="start.vad_thresh" type="integer" default="400" toc={true}>
        Voice activity detection threshold. Range: `0` to `1800`.
      </ParamField>

      <ParamField path="start.debug_level" type="integer" default="0" toc={true}>
        Debug level for logging.
      </ParamField>

      <ParamField path="start.direction" type="string[]" required={true} toc={true}>
        The direction of the call that should be transcribed. Possible values: `remote-caller`, `local-caller`.
      </ParamField>

      <ParamField path="start.speech_engine" type="string" default="deepgram" toc={true}>
        The speech recognition engine to use. Possible values: `deepgram`, `google`.
      </ParamField>

      <ParamField path="start.ai_summary_prompt" type="string" toc={true}>
        The AI prompt that instructs how to summarize the conversation when `ai_summary` is enabled.
        This prompt is sent to an AI model to guide how it generates the summary.

        Example: "Summarize the key points and action items from this conversation."
      </ParamField>
    </Indent>
  </Tab>

  <Tab title="stop">
    Set `action` to the string `"stop"` to stop the live transcription session.

    This action is designed for use on active calls that have an existing transcription session running.
    You can send this action via the [Call Commands REST API](/docs/apis/calling/calls/call-commands),
    or include it in a SWML section executed via [`transfer`](/docs/swml/reference/transfer) or [`execute`](/docs/swml/reference/execute) during a call.
  </Tab>

  <Tab title="summarize">
    <ParamField path="action.summarize" type="object" required={true} toc={true}>
      <Tip title="ai_summary vs summarize">
        If you want automatic summarization when the session ends instead, use `ai_summary: true` in the `start` action.
      </Tip>

      Request an on-demand AI summary of the conversation while transcription is still active.
      This action is designed for use on active calls that have an existing transcription session running.
      You can send this action via the [Call Commands REST API](/docs/apis/calling/calls/call-commands),
      or include it in a SWML section executed via [`transfer`](/docs/swml/reference/transfer) or [`execute`](/docs/swml/reference/execute) during a call.
      If called with an empty object, the default summarization prompt and webhook will be used.
    </ParamField>

    <Indent>
      <ParamField path="summarize.webhook" type="string" toc={true}>
        The webhook URI to be called. Authentication can also be set in the URL in the format of `username:password@url`.
      </ParamField>

      <ParamField path="summarize.prompt" type="string" toc={true}>
        The AI prompt that instructs the AI model how to summarize the conversation.
        This guides the style and content of the generated summary.

        Example: "Provide a bullet-point summary of the main topics discussed."
      </ParamField>
    </Indent>
  </Tab>
</Tabs>

## **Examples**

<Tabs>
  <Tab title="Start">
    <CodeBlocks>
      <CodeBlock title="YAML">
        ```yaml
        version: 1.0.0
        sections:
          main:
            - answer: {}
            - live_transcribe:
                action:
                  start:
                    webhook: 'https://example.com/webhook'
                    lang: en
                    live_events: true
                    direction:
                      - remote-caller
                      - local-caller
                    speech_engine: deepgram
        ```
      </CodeBlock>

      <CodeBlock title="JSON">
        ```json
        {
          "version": "1.0.0",
          "sections": {
            "main": [
              { "answer": {} },
              {
                "live_transcribe": {
                  "action": {
                    "start": {
                      "webhook": "https://example.com/webhook",
                      "lang": "en",
                      "live_events": true,
                      "direction": ["remote-caller", "local-caller"],
                      "speech_engine": "deepgram"
                    }
                  }
                }
              }
            ]
          }
        }
        ```
      </CodeBlock>
    </CodeBlocks>
  </Tab>

  <Tab title="Stop">
    <CodeBlocks>
      <CodeBlock title="YAML">
        ```yaml
        live_transcribe:
          action: stop
        ```
      </CodeBlock>

      <CodeBlock title="JSON">
        ```json
        {
          "live_transcribe": {
            "action": "stop"
          }
        }
        ```
      </CodeBlock>
    </CodeBlocks>
  </Tab>

  <Tab title="Summarize">
    <CodeBlocks>
      <CodeBlock title="YAML">
        ```yaml
        live_transcribe:
          action:
            summarize:
              webhook: 'https://example.com/webhook'
              prompt: Summarize the key points of this conversation.
        ```
      </CodeBlock>

      <CodeBlock title="JSON">
        ```json
        {
          "live_transcribe": {
            "action": {
              "summarize": {
                "webhook": "https://example.com/webhook",
                "prompt": "Summarize the key points of this conversation."
              }
            }
          }
        }
        ```
      </CodeBlock>
    </CodeBlocks>
  </Tab>
</Tabs>