AudioGen

Speech to Text API

Send an audio file URL, get a transcript with word-level timestamps. Whisper large-v3 on our own GPUs, $0.0047 per request.

Get API key See pricing

$0.0047 per request
Whisper large-v3 on our own GPUs
Unlimited on the $149 plan

Turn audio into text with one request

Send the URL of an audio file to POST /api/v6/voice/speech_to_text. The response is the transcript, with word-level timestamps when you ask for them.

Each request costs $0.0047 from your plan allowance. On the $149 Open Source Unlimited plan there is no per-request charge. One subscription and one API key cover image, video, speech and LLM APIs.

Transcription API with word timestamps

Set timestamp_level to word or sentence. You get start and end times with the text, which you can turn into SRT or VTT captions.

For long recordings, pass a webhook URL. We send the finished transcript to it.

Field	Type	What it does
init_audio	URL (required)	The mp3, wav, flac or opus file to transcribe
language	string (optional)	The language of the audio
timestamp_level	word or sentence	Adds timings to the transcript
webhook	URL (optional)	Receives the finished transcript
track_id	number (optional)	Your own reference id, sent back

Built on Whisper large-v3

Transcription runs on Whisper large-v3, the open-weight speech recognition model. We run it on our own GPUs, so you do not manage a GPU, a queue or a deployment.

For provider prices and limits side by side, see our Whisper API page.

Speech to text pricing

A transcription request costs $0.0047. On Basic ($21 a month) and Standard ($47 a month) it comes out of the dollar allowance your plan includes. On Open Source Unlimited ($149 a month) there is no per-request charge, because the model runs on our own GPUs.

Plan	Price per month	Speech to text	Parallel generations
Basic	$21	$0.0047 per request	5
Standard	$47	$0.0047 per request	10
Open Source Unlimited	$149	No per-request charge	15

Creating an account is free. API calls need a plan.

Speech to text API quick start

One POST with an audio URL. Python, JavaScript and cURL.

Python: transcribe with word timestamps

Python

1import requests
2
3response = requests.post(
4    "https://modelslab.com/api/v6/voice/speech_to_text",
5    json={
6        "key": "YOUR_MODELSLAB_API_KEY",
7        "init_audio": "https://example.com/interview.mp3",
8        "timestamp_level": "word",
9    },
10)
11
12print(response.json())

JavaScript: long file with a webhook

JavaScript

1const response = await fetch(
2  'https://modelslab.com/api/v6/voice/speech_to_text',
3  {
4    method: 'POST',
5    headers: { 'Content-Type': 'application/json' },
6    body: JSON.stringify({
7      key: 'YOUR_MODELSLAB_API_KEY',
8      init_audio: 'https://example.com/lecture.mp3',
9      timestamp_level: 'word',
10      // Long recordings queue; the webhook receives the transcript.
11      webhook: 'https://your-app.example.com/hooks/transcript',
12    }),
13  },
14);
15
16console.log(await response.json());

cURL: quick test from the terminal

bash

1curl -X POST 'https://modelslab.com/api/v6/voice/speech_to_text' \
2  -H 'Content-Type: application/json' \
3  -d '{
4    "key": "YOUR_MODELSLAB_API_KEY",
5    "init_audio": "https://example.com/voicemail.wav",
6    "timestamp_level": "word"
7  }'

Transcribe, translate and re-voice

Send the transcript to an LLM on the same key to translate or summarize it. LLMs are partner models billed per million tokens: from the plan allowance first on Basic and Standard, then the wallet, and from the wallet on the $149 plan. Then read the result back with our Text to Speech API in the voice you choose.

Speech to text by language

Language pages with transcription examples for each language.

Pricing That's Perfect

Choose plan as per your needs, cancel anytime.

100% refund policy on monthly & yearly plans — cancel anytime

Contact Sales

Best Value

Open Source Unlimited

Mission-Critical

$149/month

100% refund policy · cancel anytime

Unlimited Open Source Models

100% refund policy

24x7 Support

15 parallel generations

Access to all APIs

Unlimited generations on all open-source models

For mission critical workloads

Add Team Members

Priority GPU Clusters

Standard

Production

$47/month

100% refund policy · cancel anytime

Moderate Traffic

100% refund policy

Priority Developer Support

10 concurrent API requests

For Production workloads

API access to all models

Prototype

Basic

Prototype

$21/month

100% refund policy · cancel anytime

Moderate Traffic

100% refund policy

Developer Support via Discord/Email

5 concurrent API requests

API access to all models

Shared GPU

Start transcribing audio

Plans start at $21 a month. Each transcript is $0.0047, or unlimited on the $149 plan.

Get API key

Whisper API
Whisper large-v3 on our own GPUs.
Text to Speech API
48 languages, billed per second of audio.
Voice Cloning API
Clone a voice from a short sample.
Cheapest text to speech API
TTS cost per minute of audio, compared.
Cheapest AI API
Image, video and speech prices side by side.
Unlimited AI API
Every self-hosted open-source model for $149/mo.

Get Expert Support in Seconds

We're Here to Help.

Want to know more? You can email us anytime at support@modelslab.com

View Docs

POST https://modelslab.com/api/v6/voice/speech_to_text with `init_audio` set to the URL of the audio file. The response carries the transcript, or an id to fetch when the file is long enough to queue. Add `language` when you know it and `timestamp_level` when you need word timings.

Whisper large-v3, running on ModelsLab's own GPUs. The same handler is registered at /api/v6/voice/speech_to_text and /api/v6/whisper/transcribe.

Transcription costs $0.0047 per request whatever the length of the audio, paid from the usage included in your plan (from $21/month). Open Source Unlimited ($149/month) has no limit on the open-weight speech models ModelsLab self-hosts. Billing is per request, not per minute of audio, so one long recording costs the same as one short one.

Yes — it is a plain HTTP POST, so `requests.post()` with the JSON body is the whole integration. There is no SDK to install and no streaming session to manage; you send a URL and read the transcript from the response.

Yes. `timestamp_level` takes `word` or `sentence`, which is what you need to build SRT or VTT output.

ModelsLab is a paid service. Plans start at $21/month. Creating an account is free, but transcription requires an active plan.

Yes. The transcript can go straight into the LLM endpoint for translation and then into /api/v6/voice/text_to_audio to speak it — including in a cloned voice — and into /api/v7/video-fusion/lip-sync if the source was video. All of it runs on one API key.

Speech to Text API

Turn audio into text with one request

Transcription API with word timestamps

Built on Whisper large-v3

Speech to text pricing

Speech to text API quick start

Python: transcribe with word timestamps

JavaScript: long file with a webhook

cURL: quick test from the terminal

Transcribe, translate and re-voice

Speech to text by language

Pricing That's Perfect

Open Source Unlimited

Standard

Basic

Start transcribing audio

Related Speech APIs

We're Here to Help.

How do I convert speech to text with the API?

Which model does the speech to text API use?

How much does the speech to text API cost?

Can I use it from Python?

Does it return timestamps for subtitles?

Is there a free speech to text API?

Can I transcribe and then translate or re-voice the audio?