---
title: "Read Audio Recording"
method: GET
path: "/v2/audio-recordings/{audio_recording_id}"
tags: ["external", "external-v2"]
---

# Read Audio Recording

`GET /v2/audio-recordings/{audio_recording_id}`

Fetch the current public transcript and metadata for a single audio recording by ID.

## Path parameters

- `audio_recording_id` string, uuid, required — UUID of the audio recording to read. The caller must be able to access this recording.

## Query parameters

- `transcript_speaker_labels` boolean — When true (default), transcript turns are prefixed with available speaker labels. Labels may be participant names, generic diarization labels such as `[Speaker 1]`, or app channel labels such as `[Speaker]` and `[Other Speakers]`. Participant names are included only when available; many transcripts use generic diarization labels or app channel labels instead. Generic labels are speaker grouping context, not verified participant identity. When false, the same transcript content is returned without speaker labels.

## Response `200`

OK

- ReadAudioRecordingResponseSchema
  - `associated_note_id` string, uuid, nullable — UUID of the note directly associated with this audio recording, if any.
  - `created_at` string, date-time, required — Creation timestamp for the audio recording in ISO 8601 format.
  - `duration_seconds` integer, nullable — Audio duration in seconds when available.
  - `id` string, uuid, required — UUID of the requested audio recording.
  - `request_id` string, required — Identifier for this API request. Useful for tracing and support.
  - `transcript` string, nullable — Transcript text for the audio recording when available. By default, transcript turns are returned with available speaker labels, including participant names, generic diarization labels such as `[Speaker 1]`, and app channel labels such as `[Speaker]` and `[Other Speakers]`. Participant names are included only when available; many transcripts use generic diarization labels or app channel labels instead. Generic labels are speaker grouping context, not verified participant identity. When `transcript_speaker_labels=false` is requested, speaker labels are omitted. Null means the recording exists but transcript data is not available yet.
  - `updated_at` string, date-time, required — Last modification timestamp for the audio recording in ISO 8601 format.

---

[API](https://skmtc.dev/mem/apis/mem-api.md) · [All operations](https://skmtc.dev/mem/apis/mem-api/llms.txt) · [OpenAPI document](https://skmtc-service-production.skmtc.workers.dev/v1/apis/mem/mem-api/revisions/ad6ae6f42a8f/schema)