Fast, encrypted Speech-to-Text API

A transcription and speech recognition API that transcribes, diarizes and summarizes any audio file over REST. Hosted in France (EU), GDPR-compliant, AES-256 encrypted, and never trained on your data.

From $0.20 per audio hour. One free transcription a day.

POST /api/v1/transcription-jobs
GET /transcriptions/{id}/transcript · 200 OK
Read full API documentation

A transcription API isolated from any AI model training

Hosted on European infrastructure with AES-256 encryption and full GDPR compliance.

EU infrastructure

All servers, processing pipelines, and storage operate within European data centers, fully compliant with GDPR.

AES-256 encryption

Data is encrypted at rest across all storage layers, from upload to retrieval.

Zero model training

Audio files and output transcripts are never used to train internal or third-party AI models.

You control deletion

Transcripts and audio stay until you delete them, from the app or with one API call. Working audio used during processing is deleted after a day.

Need legal documentation for your compliance team?

Access our DPA →

Powered by the Vook engine

The audio transcription API runs on the same engine as the Vook.ai app.

Direct file ingestion

Up to 99% accuracy on clear audio. Send raw audio directly without preprocessing or output cleanup steps.

100+ languages, detected automatically

Pass a language code, or send the file and let the API detect the spoken language.

Speaker identification

Partition audio streams by speaker with exact start and end timestamps on every transcript.

AI summaries

Get a detailed summary of any transcript, in the language of the audio.

Flexible export formats

Receive structured JSON alongside pre-formatted SRT, PDF, Word, or Markdown files.

Batch & webhooks

Submit large audio files asynchronously and receive an immediate HTTP callback when processing finishes.

Integration in 3 steps

1

Get your API key

Create an account to generate your key, then try the API with one free transcription per day before buying credit.

2

Upload the file

Call the init endpoint, PUT the file to the signed URL, and trigger the transcription job.

3

Retrieve the transcript

Poll the job status or handle the webhook event to receive your formatted output.

Use Vook.ai from your AI via MCP

The Vook.ai MCP connects Claude Code, Claude Desktop, Cursor and other AI tools to the API. Same API key, same credits, no extra plan.

  • Transcribe a local file or a URL from the conversation
  • Read and summarize any transcript
  • Export to SRT, Word, PDF or Markdown
Claude Code · one line
claude mcp add vook \
  --env VOOK_API_KEY=vk_live_… \
  -- npx -y vook-mcp

# Claude Desktop, Cursor: add "npx -y vook-mcp" to your MCP config
Set-up guide for Claude Desktop, Cursor and more

Ready to build?

Sign up, generate an API key, and try it with a free transcription before buying credit.

Need high volume, custom features or a DPA? Talk to sales

Transcription API: frequently asked questions

Can't find your question? Send us a message!

Is this a speech recognition API?

Yes. Vook.ai's speech recognition API converts speech from any audio or video file to text in 100+ languages, with speaker identification and timestamps.

What audio and video formats does the API support?

The same formats as the app: common audio (MP3, WAV, M4A, FLAC, OGG, OPUS, AAC) and video (MP4, MOV, MKV, WEBM). Files up to 5 hours long are supported.

Which languages does the API support?

The API transcribes 100+ languages. Pass 'auto' as the language to detect the spoken language automatically, or set a specific language code (like 'en' or 'fr') to force it.

How accurate is the transcription?

Vook.ai reaches up to 99% accuracy on clean audio and guarantees at least 90% across all transcriptions, with automatic speaker identification and timestamps on every result.

What does the API return?

A structured JSON transcript with speaker labels and timestamps on every segment. You can also export to SRT, Word, PDF, Markdown and HTML from the same endpoint.

Does the API support speaker diarization and timestamps?

Yes. With diarize on, every segment of the JSON transcript carries a speaker key and its timestamps. Timestamps are per segment, not per word.

Can the API summarize a transcript?

Yes. Once a transcription is complete, a single call to the summary endpoint returns a Markdown summary in the transcript's language. Later calls return the same summary.

Can I use Vook.ai from Claude or Cursor?

Yes. Install vook-mcp with npx and add your API key. Your AI can then transcribe files, read and summarize transcripts, and export them. Every call is a regular API call, billed on your API credits.

How fast is transcription?

A one-hour file is transcribed in about a minute. High volumes are processed asynchronously and each job carries its own status.

Where is my audio stored, and is it secure?

Your audio and transcripts are stored in France (EU), encrypted with AES-256 at rest, and never used to train models. Nobody, not even the Vook.ai team, can access your content. They stay until you delete them through the API, and working audio is deleted after one day.

Where can I find the documentation?

Read the REST API reference for full endpoint schemas, payload structures, and runnable examples.

What does the transcription API cost?

Standard pricing for the transcription API is $0.60 per hour of audio, transcription and diarization included. For à la carte features or high volume, custom plans start from $0.20 per hour. Talk to our team for a quote. We also build custom features for API customers: tell our team what you need.

Can I test the transcription API for free?

Yes. You get one free transcription a day through the API and the MCP, with no duration cap and no export lock. You need an account and an API key. Buying API credit ends the free daily transcription.