Elevenlabs Scribe V2 API Documentation

Elevenlabs Scribe V2 API Documentation

Playground

ElevenLabs Scribe V2 Speech-to-Text transcribes audio with automatic language detection, speaker labels, and word-level timestamps for transcription, subtitles, captions, meetings, and audio processing workflows. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.

Features

ElevenLabs Scribe V2 transcribes audio into structured text with automatic language detection, optional speaker diarization, word-level timestamps, and audio-event tagging. It is designed for interviews, meetings, podcasts, voice recordings, and other speech-to-text workflows where timing and speaker information matter.

Use optional keyterms to provide recognition hints for names, technical terms, brands, or other vocabulary that may be difficult to recognize from audio alone.

Why Choose This?

• Speech-to-text transcription
  Convert recorded speech into structured transcript data.

• Automatic language detection
  Let the model detect the spoken language automatically, or provide a language code when known.

• Speaker diarization
  Enable diarize to distinguish different speakers in multi-speaker recordings.

• Word-level timestamps
  Receive timing information for individual words for subtitles, editing, alignment, and search workflows.

• Audio-event tagging
  Detect non-speech events such as laughter when tag_audio_events is enabled.

• Keyterm guidance
  Provide up to 100 recognition hints for names, brands, technical vocabulary, or domain-specific terms.

Parameters

How to Use

1. Provide the audio — Upload a recording or supply a direct public audio-file URL.
2. Set language optional — Provide language_code when the spoken language is known, or omit it for automatic detection.
3. Enable diarization optional — Turn on diarize when multiple speakers should be identified separately.
4. Configure audio events optional — Keep tag_audio_events enabled when non-speech events should be included.
5. Add keyterms optional — Supply names, brands, technical vocabulary, or other recognition hints when needed.
6. Submit — Process the recording and retrieve the transcript result.

Pricing

Pricing is based on the processed input audio duration.

Audio duration is charged proportionally rather than rounded up to full minutes.

Example Costs

Only the first 60 minutes of input audio are processed and billed.

Providing one or more keyterms applies a 1.2× multiplier to the transcription price. The number of keyterms does not add separate per-item charges.

language_code, diarize, and tag_audio_events do not add separate charges.

Best Use Cases

• Interviews — Transcribe conversations with optional speaker identification.
• Meetings — Create searchable transcripts with speaker labels and word timing.
• Podcasts — Convert spoken episodes into text for publishing, indexing, or subtitle workflows.
• Subtitle preparation — Use word-level timestamps to build caption and subtitle pipelines.
• Technical recordings — Use keyterms to improve recognition of specialized terminology.
• Content search and indexing — Convert audio archives into structured text for downstream processing.

Pro Tips

• Enable diarize when multiple people speak in the same recording.
• Provide language_code when you already know the language and want to avoid relying on automatic detection.
• Use keyterms for uncommon names, brands, acronyms, product names, or technical vocabulary.
• Keep keyterms short and specific instead of using full explanatory sentences.
• Use a clean recording with clear speech for stronger transcription results.
• Supply a direct audio-file URL rather than a webpage containing an embedded audio player.

Notes

• audio is required.
• Only the first 60 minutes of input audio are processed and billed.
• keyterms supports up to 100 items.
• Each keyterm must be under 50 characters and contain no more than 5 words.
• Keyterms provide recognition guidance but do not guarantee exact spelling.
• The result includes transcript text, detected language information, and word-level timing data.
• When enabled, diarization adds speaker labels to the transcription result.
• Public URLs must point directly to an accessible audio file.

Related Models

• ElevenLabs Forced Alignment — Align an existing transcript with audio and return detailed timing information.

Authentication

For authentication details, please refer to the Authentication Guide.

API Endpoints

Submit Task & Query Result

set -euo pipefail

export WAVESPEED_API_KEY="your-api-key"

REQUEST_BODY=$(cat <<'JSON'
{
  "audio": "https://interactive-examples.mdn.mozilla.net/media/cc0-audio/t-rex-roar.mp3",
  "diarize": false,
  "tag_audio_events": true,
  "keyterms": []
}
JSON
)

# 1. Submit the prediction.
SUBMIT_RESPONSE=$(curl --silent --show-error --fail-with-body \
  -X POST "https://api.wavespeed.ai/api/v3/elevenlabs/scribe-v2" \
  -H "Authorization: Bearer ${WAVESPEED_API_KEY}" \
  -H "Content-Type: application/json" \
  -d "${REQUEST_BODY}")

TASK=$(printf '%s' "${SUBMIT_RESPONSE}" | jq 'if type == "object" and has("data") then .data else . end')
PREDICTION_ID=$(printf '%s' "${TASK}" | jq -r '.id // empty')
if [ -z "${PREDICTION_ID}" ]; then
  printf 'Submission response did not contain a prediction id
' >&2
  exit 1
fi
RESULT_URL="https://api.wavespeed.ai/api/v3/predictions/${PREDICTION_ID}/result"

# 2. Poll until the prediction finishes.
while true; do
  RESPONSE=$(curl --silent --show-error --fail-with-body \
    "${RESULT_URL}" \
    -H "Authorization: Bearer ${WAVESPEED_API_KEY}")
  RESULT=$(printf '%s' "${RESPONSE}" | jq 'if type == "object" and has("data") then .data else . end')
  STATUS=$(printf '%s' "${RESULT}" | jq -r '.status // empty')

  case "${STATUS}" in
    completed) printf '%s\n' "${RESULT}" | jq '.outputs'; break ;;
    failed|cancelled|timeout|deleted) printf '%s\n' "${RESULT}" | jq . >&2; exit 1 ;;
    *) sleep 2 ;;
  esac
done

Parameters

Task Submission Parameters

Request Parameters

Response Parameters

Result Request Parameters

Result Response Parameters