Up and running in minutes
Drop-in compatible with the OpenAI Whisper API. Point your existing code at deeting.ai and you're done.
curl https://deeting.ai/api/v1/audio/transcriptions \
-H "Authorization: Bearer dt_live_xxxxxxxx" \
-F "file=@meeting.mp3" \
-F "response_format=verbose_json"import requests
resp = requests.post(
"https://deeting.ai/api/v1/audio/transcriptions",
headers={"Authorization": "Bearer dt_live_xxxxxxxx"},
files={"file": open("meeting.mp3", "rb")},
data={"response_format": "srt"},
)
print(resp.text)const form = new FormData();
form.append("file", fileInput.files[0]);
form.append("response_format", "verbose_json");
const resp = await fetch("https://deeting.ai/api/v1/audio/transcriptions", {
method: "POST",
headers: { Authorization: "Bearer dt_live_xxxxxxxx" },
body: form,
});
const data = await resp.json();Built for developers, priced for makers
Production-grade transcription without the enterprise price tag.
60ร realtime speed
A 1-hour podcast transcribes in about a minute. Batched GPU inference keeps throughput high even at scale.
OpenAI-compatible
Drop-in replacement for the OpenAI Whisper API. Change the base URL and your existing code just works.
Every output format
text, json, verbose_json, srt, and vtt โ plus word-level timestamps for karaoke-style captions.
99+ languages
Auto language detection across 99+ languages with strong multilingual and code-switching accuracy.
Simple, honest pricing
Start free with 10 hours a month. Pay a flat monthly plan that includes hours โ no surprise overages.
Any audio or video
mp3, mp4, wav, m4a and more. Upload files directly or pass an audio_url for large files.
How it works
1. Grab your API key
Sign up in seconds โ your key is issued instantly. No credit card required for the free tier.
2. Make one request
POST a file or an audio_url with your key. It's one endpoint, OpenAI-compatible.
3. Get clean text
Receive transcripts with timestamps in the format you need โ text, JSON, or subtitles.
Simple, honest pricing
Free to start. Scales with your product.
All plans include every output format, word timestamps, and 99+ languages.
Frequently asked questions
โบIs this really OpenAI API compatible?
Yes. The endpoint mirrors OpenAI's /v1/audio/transcriptions request and response shapes, so existing OpenAI Whisper integrations work by changing only the base URL.
โบHow accurate is the transcription?
We run Whisper large-v3, the most accurate open-source speech model, with int8 GPU quantization for speed without meaningful quality loss.
โบWhat happens when I run out of free hours?
Requests return a clear 429 quota_exceeded error. Upgrade to a paid plan anytime โ your included hours reset monthly.
โบDo you store my audio?
No. Audio is transcribed in memory and discarded immediately. We only record audio duration for billing, never content.
Start transcribing in the next 60 seconds
10 free hours a month. No credit card. Your API key is ready the moment you sign up.
Get your free API key