Tools

Video Transcription API

Submit any video file. Get back a full transcript with timestamps. MP4, MOV, WebM, and more.

Try it below:

FILEDrop a file or click to upload

Production-ready

Private by design

Security-first

Scalable

Production-ready

Private by design

Security-first

Scalable

Production-ready

Private by design

Security-first

Scalable

Transcribe video files

Upload a video file or submit a URL, get a transcript back.

{
  "items": [
    {
      "sequence": 1,
      "timeStart": 0.0,
      "timeEnd": 12.4,
      "text": "Welcome to the engineering onboarding series. This video covers your local development setup, how we handle deployments, and where to find documentation for each service."
    },
    {
      "sequence": 2,
      "timeStart": 12.4,
      "timeEnd": 28.7,
      "text": "Let's start with your dev environment. You'll need Node 20 or later and Docker Desktop. Clone the monorepo from GitHub and run the bootstrap script. It pulls all service dependencies and spins up the local stack."
    },
    {
      "sequence": 3,
      "timeStart": 28.7,
      "timeEnd": 41.3,
      "text": "The bootstrap takes about three minutes the first time. Once it's done you should see all six services running in Docker. Hit localhost 3000 to confirm the dashboard loads."
    },
    {
      "sequence": 4,
      "timeStart": 41.3,
      "timeEnd": 55.9,
      "text": "For deployments, we use GitHub Actions. Every merge to main triggers a build, runs the test suite, and deploys to staging automatically. Production deploys require a manual approval step in the Actions UI."
    }
  ],
  "total": 48,
  "start": 1,
  "end": 4
}

What is Exabase's video transcription API?

Submit a video file to the Exabase Extract API and get a transcript back. The API extracts the audio track, transcribes it, and returns the full transcript text alongside timestamped chunks. Each chunk carries timeStart and timeEnd in seconds, so you can map any piece of text back to the exact moment in the video. MP4, MOV, AVI, WebM, MKV, and other major formats are supported.

The same POST /v2/extract endpoint accepts both file uploads and URLs pointing to hosted video. Transcription runs on Exabase's infrastructure. No FFmpeg for demuxing, no speech-to-text model to deploy, no GPU to provision.

What you get back

The response includes the full transcript in extraction.media.transcript and the transcript split into chunks. Each chunk carries a timeStart, timeEnd, and the transcript text for that segment. You also get video duration, frame dimensions (width and height), MIME type, file size, total chunk count, and a CDN-hosted thumbnail. The API reference has the complete response schema.

One multi-modal API

The same POST /v2/extract endpoint handles every content type Exabase supports. For video, it extracts the audio track and returns timestamped transcript chunks alongside frame dimensions. For audio files, it does the same without the visual metadata. For PDFs, it returns text chunks with page numbers and (for recognised document types) structured field extraction. For images, it runs OCR. For web pages, it renders JavaScript and extracts the content.

The submission flow is identical across all types: one endpoint, one SDK method, one webhook configuration. You learn the pattern once and use it for everything. The Extract docs cover each content type in detail.

What you can build with it

Transcribe recorded presentations and make every slide's narration searchable. Build a video library where users find the right clip by searching spoken content instead of scrubbing through footage. Generate subtitle files from the timestamped chunks. Process training videos and let employees search for specific instructions by keyword.

Store transcribed video as Resources in a Base and the transcript becomes searchable through Deep Search at the chunk level. Feed timestamped chunks into a RAG pipeline so your agent can cite the exact moment in a video. Workers can process new recordings as they arrive, keeping your transcript index current without manual work.

Beyond transcription

The Extract API returns more than a transcript. You get video duration, frame dimensions, MIME type, a CDN-hosted thumbnail, and searchable chunks. Store the output as a Resource and it's searchable through Deep Search immediately. Extract key details into Memory so your agent retains what it saw between sessions. The Submitting Jobs guide at https://exabase.io/docs/developer-guide/extraction/submitting-jobs walks through the full flow.

How do I use it?

  1. Get your API key

Free, no credit card:

Sign up at exabase.io and copy your API key from the dashboard.

  1. Submit a video file

Using the SDK (Node.js):

import { Exabase } from "@exabase/sdk";
import { createReadStream } from "fs";

const api = new Exabase({
  apiKey: process.env.EXABASE_API_KEY,
});

const job = await api.extract.createFromFile({
  file: createReadStream("./onboarding-walkthrough.mp4"),
  name: "New hire onboarding — engineering",
});

console.log(job.id);    // extraction job id
console.log(job.state); // "pending"
  1. Get the result

Poll the job until state reaches completed, or skip polling entirely with webhooks.

const result = await api.extract.get({ jobId: job.id });

if (result.state === "completed") {
  console.log(result.extraction?.media?.transcript); // full transcript
  console.log(result.extraction?.media?.duration);   // duration in seconds

  const chunks = await api.extract.getChunks({
    jobId: job.id,
    start: 1,
    end: 20,
  });

  for (const chunk of chunks.items) {
    console.log(chunk.timeStart, chunk.timeEnd, chunk.text);
  }
}

One API call. No FFmpeg, no speech-to-text model, no GPU.

What you get back

Extraction job

Field

Type

Description

state

string

completed or failed

kind

string

video

extraction.common.mimeType

string

Detected MIME type (e.g. video/mp4)

extraction.common.size

number

File size in bytes

extraction.common.thumbnail

string

CDN-hosted thumbnail URL

extraction.common.chunkCount

number

Number of transcript chunks

extraction.media.duration

number

Duration in seconds

extraction.media.width

number

Frame width in pixels

extraction.media.height

number

Frame height in pixels

extraction.media.transcript

string

Full transcript text

Transcript chunks

Field

Type

Description

sequence

number

Chunk position in the transcript

text

string

Transcript text for this segment

timeStart

number

Start time in seconds

timeEnd

number

End time in seconds

Chunks are paginated. Pass start and end to GET /v2/extract/{jobId}/chunks to page through them.

API quick reference

Endpoint

Method

Description

/v2/extract

POST

Submit extraction job

/v2/extract

GET

List extraction jobs

/v2/extract/{jobId}

GET

Poll for result

/v2/extract/{jobId}/chunks

GET

Fetch transcript chunks

/v2/extract-settings

PUT

Configure webhook

Auth: X-Api-Key header on all requests.

File retention: Stored files are retained for 1 day from job creation. Download or copy anything you need before they expire.

Need file extraction or full page content?

Exabase offers a full set of tools to extract content from any media, document or website.

All your data extraction needs in one place.

Why Exabase

Comparisons

Use cases

FAQs

What is Exabase?

Exabase is infrastructure for AI agents. It gives your agents memory, versioned file storage, AI deep search, and context automation through a set of APIs. Store what your agent learns, search inside any content type, and keep knowledge bases current automatically. Built for production use. Give your agent precise context and cut your token spend by up to 81%.

Who uses Exabase?

Developers and teams building AI agents, copilots, and RAG applications. If your agent needs to remember things between sessions, store and retrieve files, search across documents and media, or stay up to date without manual maintenance, Exabase handles that infrastructure so you can focus on your product.

How fast is transcription?

Processing time depends on the file length, resolution, and format. Processing is asynchronous, so your application is not blocked while transcription runs. Poll for results or configure a webhook to be notified on completion.

What video formats are supported?

MP4, MOV, AVI, WebM, MKV, and other major video formats. The API auto-detects the content type and processes accordingly.

Does it transcribe on-screen text as well as speech?

The API transcribes the audio track. It does not perform OCR on video frames. For text visible on screen (slides, captions burned into the video), the spoken narration is captured but the on-screen text is not extracted separately.

Does it support speaker diarisation?

The API returns a full transcript split into timestamped chunks. Speaker diarisation (identifying who said what) is not currently a separate output field.

Can I submit a URL instead of uploading a file?

Yes. Pass a url field in the request body pointing to a publicly accessible video file. Exabase fetches and processes it.

Do I have to poll for results?

No. You can configure a webhook URL and Exabase will POST the full result to your server when processing completes. The webhooks guide covers setup.

Is there a file length limit?

Processing is asynchronous with no hard recording-length limit. Long files take proportionally longer to process.

Does it extract visual content from video frames?

The API extracts the audio track and transcribes it. It returns frame dimensions and a thumbnail, but does not perform frame-by-frame visual analysis or object detection. We are working on this capability.

What happens if transcription fails?

The job state moves to failed. You can call the reprocess endpoint to retry without re-uploading the original file.

How long are files retained?

Stored files are retained for 1 day from job creation, then permanently deleted. Download or copy anything you need before the retention window expires.

How is this different from the Video to JSON tool?

The Video to JSON page covers the full extraction response: metadata, duration, dimensions, thumbnail, transcript, and chunks. Video Transcription focuses specifically on the transcript and timestamped chunks. Same underlying API, different emphasis.

How is this different from the Audio Transcription tool?

Both use the same POST /v2/extract endpoint and return timestamped transcript chunks. The difference is the input: Video Transcription accepts video files (MP4, MOV, etc.) and extracts the audio track before transcribing. Audio Transcription accepts audio files (MP3, WAV, etc.) directly.

Can I use the transcript with other Exabase features?

Yes. Store transcribed video as a Resource in a Base and the transcript becomes searchable through Deep Search. Extract key details into Memory so your agent retains what it saw. Workers can process new recordings on a schedule.

Is there an SDK?

Yes. The @exabase/sdk package for Node.js/TypeScript handles file streaming, job polling, and chunk retrieval. Install with npm install @exabase/sdk. Or call the REST API directly from any language.

What other content types does Extract support?

PDFs, images (with OCR), audio files (with transcription), web pages, and documents. The same POST /v2/extract endpoint handles all of them. See the PDF to JSON, Image to Text, Audio Transcription, and Website to JSON tool pages.

Deciding?

Ask your favourite AI about Exabase:

Cut your token spend and give your agent precise context.

Get started in minutes.

Cut your token spend and give your agent precise context.

Get started in minutes.