Production-ready speech-to-text API

Transcription that moves as fast as your product.

Handle single-file transcription, async jobs, massive batches, polling, and signed webhooks with one API built for reliable transcription at scale.

$0.15 per audio hour 99 languages
api.transcribeapi.com/v1/transcribe
$ curl -X POST https://api.transcribeapi.com/v1/transcribe \ -H "Authorization: Bearer YOUR_API_KEY" \ -F "file=@meeting.mp3" \ -F "webhook_url=https://app.com/webhook" upload accepted job_8f41c2 processing...
Live job transcribing
"text": "Your transcript appears here..."
01 Upload audio multipart or JSON
02 Process job sync, async, batch
03 Return text polling or webhook
$0

Per audio hour. Billed per second, rounded up to the nearest second.

0K

Files per batch, with up to 10 GB total batch size.

0min

Free transcription for every new account.

0

Languages supported by Whisper-family speech recognition.

One clean workflow

From audio file to transcript without glue code.

A simple API surface for quick uploads, long-running jobs, batch processing, and webhook delivery.

Send audio

Upload a single file directly, or submit a batch job with thousands of files.

{ }

Choose mode

Use sync for short audio, async for large files, and batch jobs for scale.

Track status

Poll job status or receive signed webhook events when transcription completes.

Use transcript

Receive structured text ready for search, analytics, AI workflows, or product features.

Developer-first API

Copy, paste, transcribe.

Keep the homepage focused on what developers care about: clear requests, predictable responses, and a fast path to production.

curl -X POST https://api.transcribeapi.com/v1/transcribe \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -F "file=@audio.mp3" \
  -F "language=en" \
  -F "webhook_url=https://app.com/webhook"import { TranscribeAPI } from "@transcribeapi/sdk";

const client = new TranscribeAPI({
  apiKey: process.env.TRANSCRIBE_API_KEY
});

const job = await client.transcribe({
  file: "./audio.mp3",
  language: "en",
  webhookUrl: "https://app.com/webhook"
});

console.log(job.id, job.status);from transcribeapi import TranscribeAPI

client = TranscribeAPI(api_key="YOUR_API_KEY")

job = client.transcribe(
    file="audio.mp3",
    language="en",
    webhook_url="https://app.com/webhook"
)

print(job.id, job.status)
Built for teams that ship

Powerful infrastructure, simple product experience.

Simple REST API

One API for single files, async jobs, and batch processing. POST multipart or JSON and move fast.

Fast batches

Process high-volume audio workloads without waiting a full day for results.

SDK

SDKs for every stack

Use Node.js, Python, and Cloudflare Workers SDKs with uploads, polling, and retries handled.

Webhooks & polling

Get signed webhook events when jobs finish, or poll status from your backend or serverless functions.

Privacy first

Audio files are deleted immediately after transcription, keeping the flow clean and predictable.

DOC

Detailed docs

Every endpoint, error code, option, and common workflow documented with copy-paste examples.

Predictable pricing

Make the low price impossible to miss.

Use this calculator on the page so visitors instantly understand what $0.15 per audio hour means for their own volume.

1 hour 10,000 hours

Estimated monthly transcription cost

$150.00/ month

1,000 audio hours × $0.15/hour

$0.0025 10 minutes
$0.15 1 hour
$1.50 10 hours
Fast transcription at any scale

Show speed visually, not only in a table.

Completion-time expectations become easier to trust when they look like a performance chart.

Under 10 sec
1–4 sec
Under 10 min
Up to 25 sec
Single file < 1 hr
Up to 5 min
Single file 1–5 hrs
Up to 15 min
Single file 5–10 hrs
Up to 30 min
Powered by open-source Whisper

Pick speed or maximum accuracy.

Present the models like a product choice so developers understand when to use each one.

Fast default

Whisper Large v3 Turbo

Optimized for speed and near real-time transcription. Ideal for interactive applications, quick turnaround, and high-volume workflows.

  • Low latency
  • Great for product UX
  • Excellent throughput
Maximum accuracy

Whisper Large v3

Best for challenging audio, accents, background noise, multilingual content, and accuracy-sensitive workflows.

  • Handles noisy audio
  • Strong multilingual support
  • Best for difficult files
99 languages

Global speech recognition for global products.

English Spanish Hindi French German Arabic Japanese Korean Portuguese Italian Dutch Turkish Vietnamese Polish Swedish English Spanish Hindi French German Arabic Japanese Korean Portuguese Italian Dutch Turkish Vietnamese Polish Swedish
Ready to transcribe?

Start transcribing in under 2 minutes.

Get 60 minutes free, send one API request, and receive your first transcript without committing to a large plan.