Whisper STT: Speech to Text Transcription in 3 Minutes
A free month of testing, so you can check the quality on your own recordings.
What Is Whisper STT and How Does Speech Transcription Work?
Whisper is an AI model by OpenAI, trained on 680,000 hours of recordings. We provide it as a simple API: you send an audio file, you get text back. No queues and no per-minute limits.
- Open-source model by OpenAI
- Over 99 languages and dialects
- Automatic language detection
- Word and segment timestamps
- Audio/video files up to 1 GB
Audio Transcription: What Problems Does It Solve?
Manual transcription costs hours of work for every recording. Whisper does the same job in a few minutes.
Time Savings
An hour-long recording in about 3 minutes instead of a full day of manual work.
Cost Reduction
Transcription cost drops to a fraction of a transcriptionist's rate, and you get the result within minutes.
99+ Languages
Automatic transcription in virtually any language. No additional tools needed.
Content Search
Turn unsearchable audio into text, find any fragment in seconds.
How Does the Speech-to-Text API Work?
- 1
Send a File
Upload audio/video via API: MP3, WAV, MP4, WEBM and more.
- 2
GPU Processes
Whisper analyzes the recording on NVIDIA GPUs. One hour of audio ≈ 3 minutes.
- 3
Get Your Text
Ready transcription in your format of choice, with or without timestamps.
This Is What the Transcript Looks Like
You pick the format when you upload the file. JSON gives you segments with timestamps and the detected language, while SRT and VTT drop straight into a video player or YouTube.
{
"language": "en",
"duration": 3612.4,
"segments": [
{
"id": 0,
"start": 0.0,
"end": 4.32,
"text": "Good morning, let us start the board meeting."
},
{
"id": 1,
"start": 4.32,
"end": 9.18,
"text": "First item on the agenda is third quarter sales."
}
]
}Whisper API on NVIDIA GPUs: Why Faster Than the Cloud?
GPU, Not CPU
NVIDIA A100 and H100 cards. One hour of audio takes about 3 minutes of processing.
Data Stays in Poland
Your files never leave the country. Full GDPR compliance.
You Pick the Format
You choose the variant (Express, Standard or Precision), the format (SRT/VTT/JSON) and the language of the recording.
Straightforward integration
One REST endpoint and OpenAPI docs. We hand you the examples in Python, Node.js and cURL.
Scales With You
From a single file to thousands of recordings per day. Infrastructure grows automatically.
Real Humans
Tech support from the team that built and runs this API.
Speech Transcription: Subtitles, Minutes, Call Analysis
See What Transcription Costs
Set how much audio you process per month. You pay only for processed minutes.
Your starting price
PLN 54/ mo.PLN 108
At this volume you pay PLN 0.009 per minute of audio, in 99+ languages, with SRT/VTT/JSON export.
On AWS Transcribe the same volume costs about PLN 135 per month.
Volume tiers
volume discount -10%- first 50 hPLN 0.01 / min
- 50 to 200 hPLN 0.008 / min
- 200 to 1,000 hPLN 0.007 / min
- over 1,000 hPLN 0.006 / min
Tiers are billed separately, like tax brackets: volume that falls into a lower tier keeps that tier's rate. Rates in the table already include the -50% intro discount.
Your first month of testing is free. The -50% intro rate for new clients kicks in after it.
Order this package
Leave your contact details. The calculated volume reaches us with your inquiry, and we get back within 24h with access and details.
Test Speech Transcription for a Month, Free
Upload your own recordings and see how Whisper STT handles your material: the accents, the jargon and the audio quality.
We respond within 24h. We help with the rollout.