podcast_episode_42.mp3 — 38:17
$0
Per audio hour
Billed per second, no minimums
0
Files per batch
Up to 10 GB per batch request
0 min
Free transcription
Every new account, no card needed
0
Languages supported
Powered by OpenAI Whisper

Built for developers who need scale

Every endpoint is designed for production. Direct responses for short files, async jobs for everything else, and webhooks so you never miss a result.

Lightning Fast

Under 10 seconds of audio returns in 1-4 seconds. Files under 10 minutes complete within 25 seconds. Optimized for minimal latency at every tier.

100+ Languages

Automatic language detection or specify any of 100+ ISO 639-1 codes. From Afrikaans to Zulu, Cantonese to Yoruba — we have you covered.

Privacy by Design

Audio files deleted immediately after transcription. Results auto-expire after 7 days. Instant deletion available via API. Your data stays yours.

SDKs for Every Stack

First-class Node.js, Python, and Cloudflare Workers SDKs. Auto-reload balances, polling, batching — all handled so you focus on building.

Webhooks & Polling

Receive completed transcripts at your server with signed webhooks. Poll for status when you need it. Signature verification keeps delivery secure.

Simple Pricing

$0.15 per hour of audio. No hidden fees. Auto-reload keeps your balance topped up. 60 minutes free when you save a card. Pay as you go.

Drop-in SDKs for your stack

The SDK handles upload strategy, polling, and auto-reload so you write less code.

import { TranscribeAPI } from "@transcribe-api/sdk";

const client = new TranscribeAPI({
  apiKey: "YOUR_API_KEY",
  autoReload: { amount: 10, ifBalanceBelow: 20 },
  polling: { interval: 10, timeout: 15 * 60 },
  showLogs: true,
});

const result = await client.transcribe({
  files: [
    { reference_id: "meeting_1", file: "meeting.mp3" },
  ],
  language: "en",
  exclude: ["metadata"],
});

// Transcript returned — autoReload checked balance
console.log(result);
Speed

Fast transcription at any scale

Maximum expected completion times — most jobs finish faster.

Under 10 seconds
1–4 sec
Under 10 minutes
up to 25 sec
Single file < 1 hour
up to 5 min
Single file 1–5 hours
up to 15 min
Single file 5–10 hours
up to 30 min

Batch jobs add ~100–300 ms per file on top of the duration-based latency.

Limits

Clear limits, no surprises

Transcription Endpoints

Free tier10 RPM
Pay as you go300 RPM

File & Batch Limits

Single file (sync)30 MB / 10 min
Single file (async)10 GB / 10 hrs
Batch files10,000
Total batch size10 GB
Pricing

Simple, predictable pricing

Pay only for what you transcribe. Billed per second, rounded up to the nearest second. No setup fees, no hidden costs, no minimums.

10 minutes$0.025
1 hour$0.15
10 hours$1.50
100 hours$15.00
1,000 hours$150.00

New accounts get 60 minutes free. Save a card to prevent abuse — no charge until you exceed your free minutes.

Audio hours per month 100 hrs
$15.00
estimated monthly cost
Audio hours100 hrs
Rate$0.15 / hr
Monthly total$15.00
Per file (avg 45 min)$0.1125
Languages

99+ languages, zero config.

Set language=auto and we handle detection automatically.

// get started in under 2 minutes

Start transcribing
for free today

Get 60 free minutes. Send one API request.
Receive your transcript.

No credit card required  ·  $0.15 / hr after free tier  ·  Cancel anytime