Scribe Speech-to-Text

Transcribe audio or video with word timestamps and optional speaker labels.

by ElevenLabsReady to run$0.0052 per minute

Example

You ask your agent

“Transcribe my one-minute audio clip and label the speakers.”

You get

Text: Welcome to Scrollport. Our product meeting is in London on Monday. Alex will present the new research tools, and the…
Entities: Scrollport, London, Monday and 2 more
Language code: eng
Duration (seconds): 8
Language probability: 0.955

What your agent sends and gets back (JSON)

Your agent works out the exact inputs, such as IDs, dates and settings, from what you ask.

Request

{
  "tool_id": "elevenlabs.scribe-speech-to-text",
  "input": {
    "diarize": true,
    "keyterms": [
      "Scrollport"
    ],
    "model_id": "scribe_v2",
    "audio_url": "https://scrollport.com/examples/transcription-sample.mp3",
    "entity_detection": true,
    "expected_duration_minutes": 0.15
  }
}

Response

{
  "text": "Welcome to Scrollport. Our product meeting is in London on Monday. Alex will present the new research tools, and the team will review customer feedback",
  "words": [
    {
      "end": 0.4,
      "text": "Welcome",
      "type": "word",
      "start": 0.1,
      "speaker_id": "speaker_0"
    },
    {
      "end": 0.44,
      "text": " ",
      "type": "spacing",
      "start": 0.4,
      "speaker_id": "speaker_0"
    },
    {
      "end": 0.48,
      "text": "to",
      "type": "word",
      "start": 0.44,
      "speaker_id": "speaker_0"
    },
    {
      "end": 0.54,
      "text": " ",
      "type": "spacing",
      "start": 0.48,
      "speaker_id": "speaker_0"
    },
    {
      "end": 1,
      "text": "Scrollport.",
      "type": "word",
      "start": 0.54,
      "speaker_id": "speaker_0"
    },
    {
      "end": 1.54,
      "text": " ",
      "type": "spacing",
      "start": 1,
      "speaker_id": "speaker_0"
    },
    {
      "end": 1.66,
      "text": "Our",
      "type": "word",
      "start": 1.54,
      "speaker_id": "speaker_0"
    },
    {
      "end": 1.7,
      "text": " ",
      "type": "spacing",
      "start": 1.66,
      "speaker_id": "speaker_0"
    },
    {
      "end": 1.96,
      "text": "product",
      "type": "word",
      "start": 1.7,
      "speaker_id": "speaker_0"
    },
    {
      "end": 2,
      "text": " ",
      "type": "spacing",
      "start": 1.96,
      "speaker_id": "speaker_0"
    },
    {
      "end": 2.3,
      "text": "meeting",
      "type": "word",
      "start": 2,
      "speaker_id": "speaker_0"
    },
    {
      "end": 2.36,
      "text": " ",
      "type": "spacing",
      "start": 2.3,
      "speaker_id": "speaker_0"
    },
    {
      "end": 2.42,
      "text": "is",
      "type": "word",
      "start": 2.36,
      "speaker_id": "speaker_0"
    },
    {
      "end": 2.46,
      "text": " ",
      "type": "spacing",
      "start": 2.42,
      "speaker_id": "speaker_0"
    },
    {
      "end": 2.54,
      "text": "in",
      "type": "word",
      "start": 2.46,
      "speaker_id": "speaker_0"
    },
    {
      "end": 2.58,
      "text": " ",
      "type": "spacing",
      "start": 2.54,
      "speaker_id": "speaker_0"
    },
    {
      "end": 2.86,
      "text": "London",
      "type": "word",
      "start": 2.58,
      "speaker_id": "speaker_0"
    },
    {
      "end": 2.88,
      "text": " ",
      "type": "spacing",
      "start": 2.86,
      "speaker_id": "speaker_0"
    },
    {
      "end": 3,
      "text": "on",
      "type": "word",
      "start": 2.88,
      "speaker_id": "speaker_0"
    },
    {
      "end": 3.04,
      "text": " ",
      "type": "spacing",
      "start": 3,
      "speaker_id": "speaker_0"
    },
    {
      "end": 3.38,
      "text": "Monday.",
      "type": "word",
      "start": 3.04,
      "speaker_id": "speaker_0"
    },
    {
      "end": 3.38,
      "text": " ",
      "type": "spacing",
      "start": 3.38,
      "speaker_id": "speaker_0"
    },
    {
      "end": 4.28,
      "text": "Alex",
      "type": "word",
      "start": 4.019,
      "speaker_id": "speaker_0"
    },
    {
      "end": 4.3,
      "text": " ",
      "type": "spacing",
      "start": 4.28,
      "speaker_id": "speaker_0"
    },
    {
      "end": 4.44,
      "text": "will",
      "type": "word",
      "start": 4.3,
      "speaker_id": "speaker_0"
    },
    {
      "end": 4.46,
      "text": " ",
      "type": "spacing",
      "start": 4.44,
      "speaker_id": "speaker_0"
    },
    {
      "end": 4.76,
      "text": "present",
      "type": "word",
      "start": 4.46,
      "speaker_id": "speaker_0"
    },
    {
      "end": 4.8,
      "text": " ",
      "type": "spacing",
      "start": 4.76,
      "speaker_id": "speaker_0"
    },
    {
      "end": 4.88,
      "text": "the",
      "type": "word",
      "start": 4.8,
      "speaker_id": "speaker_0"
    },
    {
      "end": 4.94,
      "text": " ",
      "type": "spacing",
      "start": 4.88,
      "speaker_id": "speaker_0"
    },
    {
      "end": 5.08,
      "text": "new",
      "type": "word",
      "start": 4.94,
      "speaker_id": "speaker_0"
    },
    {
      "end": 5.12,
      "text": " ",
      "type": "spacing",
      "start": 5.08,
      "speaker_id": "speaker_0"
    },
    {
      "end": 5.46,
      "text": "research",
      "type": "word",
      "start": 5.12,
      "speaker_id": "speaker_0"
    },
    {
      "end": 5.54,
      "text": " ",
      "type": "spacing",
      "start": 5.46,
      "speaker_id": "speaker_0"
    },
    {
      "end": 5.76,
      "text": "tools,",
      "type": "word",
      "start": 5.54,
      "speaker_id": "speaker_0"
    },
    {
      "end": 5.82,
      "text": " ",
      "type": "spacing",
      "start": 5.76,
      "speaker_id": "speaker_0"
    },
    {
      "end": 5.9,
      "text": "and",
      "type": "word",
      "start": 5.82,
      "speaker_id": "speaker_0"
    },
    {
      "end": 5.94,
      "text": " ",
      "type": "spacing",
      "start": 5.9,
      "speaker_id": "speaker_0"
    },
    {
      "end": 6.04,
      "text": "the",
      "type": "word",
      "start": 5.94,
      "speaker_id": "speaker_0"
    },
    {
      "end": 6.1,
      "text": " ",
      "type": "spacing",
      "start": 6.04,
      "speaker_id": "speaker_0"
    },
    {
      "end": 6.32,
      "text": "team",
      "type": "word",
      "start": 6.1,
      "speaker_id": "speaker_0"
    },
    {
      "end": 6.36,
      "text": " ",
      "type": "spacing",
      "start": 6.32,
      "speaker_id": "speaker_0"
    },
    {
      "end": 6.48,
      "text": "will",
      "type": "word",
      "start": 6.36,
      "speaker_id": "speaker_0"
    },
    {
      "end": 6.5,
      "text": " ",
      "type": "spacing",
      "start": 6.48,
      "speaker_id": "speaker_0"
    },
    {
      "end": 6.74,
      "text": "review",
      "type": "word",
      "start": 6.5,
      "speaker_id": "speaker_0"
    },
    {
      "end": 6.84,
      "text": " ",
      "type": "spacing",
      "start": 6.74,
      "speaker_id": "speaker_0"
    },
    {
      "end": 7.16,
      "text": "customer",
      "type": "word",
      "start": 6.84,
      "speaker_id": "speaker_0"
    },
    {
      "end": 7.24,
      "text": " ",
      "type": "spacing",
      "start": 7.16,
      "speaker_id": "speaker_0"
    },
    {
      "end": 7.62,
      "text": "feedback",
      "type": "word",
      "start": 7.24,
      "speaker_id": "speaker_0"
    }
  ],
  "entities": [
    {
      "text": "Scrollport",
      "end_char": 21,
      "start_char": 11,
      "entity_type": "organization"
    },
    {
      "text": "London",
      "end_char": 55,
      "start_char": 49,
      "entity_type": "location_city"
    },
    {
      "text": "Monday",
      "end_char": 65,
      "start_char": 59,
      "entity_type": "date"
    },
    {
      "text": "Alex",
      "end_char": 71,
      "start_char": 67,
      "entity_type": "name"
    },
    {
      "text": "Alex",
      "end_char": 71,
      "start_char": 67,
      "entity_type": "name_given"
    }
  ],
  "language_code": "eng",
  "duration_seconds": 8,
  "language_probability": 0.9550108313560486
}

Use it with your agent

Copy the prompt and paste it into your agent, then tell it what you need.

Read https://scrollport.com/start. Use my existing Scrollport connection, or help me connect if needed. Help me use Scribe Speech-to-Text (tool ID: elevenlabs.scribe-speech-to-text). Ask me for the details you need, then show me the result.

Provider access is included. If your agent isn't connected to Scrollport yet, it will help you set that up.

Questions

Is Scribe Speech-to-Text by ElevenLabs free?

Scrollport has no free tier: it costs $0.0052 per minute, paid per use from a prepaid wallet with no subscription. Connecting your first agent adds $1 of trial credit, so you can try it before adding funds.

Do I need an API key from ElevenLabs?

No. Your agent runs it through your Scrollport connection (MCP, the Scrollport CLI or the HTTP API), and Scrollport handles access to ElevenLabs, so you don't sign up with ElevenLabs or manage an API key.

Details

About this tool

Supply a public HTTPS audio/video URL or a signed workspace-file URL and its expected duration in minutes. Scribe v2 is the default; v1 is available. Language is detected automatically unless specified. Optional speaker labels and word timestamps are included in the base rate. Keyterms, entity detection and non-verbatim mode require v2. Keyterm prompting and entity detection each select a higher per-minute rate; inspect the quote before running. Billing uses actual transcribed seconds, not processing time. Round expected_duration_minutes up if uncertain; it does not trim the file. A transcript that exceeds the quoted duration is rejected without charging your wallet. Files must be under 2 GB and no longer than 600 minutes. Character timestamps, multichannel transcription, webhooks and account-level speaker libraries are not included.

All inputs and outputs

Input

audio_urlstring · required

Public HTTPS audio/video URL or signed workspace file URL; maximum file size 2 GB.

format: uri · maxLength: 8192

diarizeboolean · optional

entity_detectionboolean · optional

Detect entities in the transcript. Adds to the per-minute rate.

expected_duration_minutesnumber · required

Audio length in minutes, rounded up if uncertain. Sizes the quote; does not trim the file. Settlement uses actual audio duration.

minimum: 0.001 · maximum: 600

keytermsarray · optional

Optional domain terms. A nonempty list adds to the per-minute rate.

maxItems: 100

language_codestring · optional

minLength: 2 · maxLength: 8

model_idstring · optional

enum: scribe_v2, scribe_v1

no_verbatimboolean · optional

num_speakersinteger · optional

minimum: 1 · maximum: 32

tag_audio_eventsboolean · optional

timestamps_granularitystring · optional

enum: none, word

Output

duration_secondsnumber

Length of the media in seconds.

entitiesarray

language_codestring

language_probabilitynumber

textstring

wordsarray

Errors and limits

audio_duration_exceeded: The audio was longer than expected_duration_minutes; the wallet was not charged.Check the full audio length and request a new quote with enough minutes.
invalid_input: Use a reachable HTTPS audio/video URL and supported options.Check file size, signed-link expiry and the expected duration.