Transcribe with Whisper Turbo
5.1%
WER
99
Languages
32.0x
Speed
MIT
License
About Whisper Turbo
Whisper Turbo (large-v3-turbo) is a distilled version of Whisper Large V3 that dramatically reduces inference time while maintaining competitive accuracy. With only 4 decoder layers instead of 32, it achieves a 4x speedup.
Model Info
- ProviderOpenAI
- Architecture-
- LicenseMIT
- UpdatedMar 2026
Frequently Asked Questions
Whisper Turbo is a speech-to-text model by OpenAI. STT.ai hosts Whisper Turbo on our GPU infrastructure so you can use it without provisioning your own hardware — upload audio or video and pick Whisper Turbo from the model picker.
On standard benchmarks, Whisper Turbo achieves around 5.1% Word Error Rate. Real-world accuracy depends on audio quality, accent, and language; for noisy or accented recordings, expect a few percentage points higher WER.
Whisper Turbo runs on STT.ai's free tier — every visitor gets 600 minutes/month at no cost. Paid plans add longer per-file limits, private transcripts, and priority queueing.
Whisper Turbo is released under MIT, a permissive open-source license. You can self-host Whisper Turbo on your own hardware or use our hosted version — both are commercially usable.
Whisper Turbo supports 99 languages. Auto-detection picks the right language for most audio; you can also specify it manually for a small accuracy lift.
Whisper Turbo processes audio at about 32.0x real-time on our GPUs. A 1-hour audio file finishes in under 1 minutes; longer files queue and notify by email when done.
Whisper Turbo has 809M parameters. Larger models tend to be more accurate but slower; STT.ai hosts Whisper Turbo on GPU so the parameter count doesn't affect your client-side performance.
Whisper Turbo accepts every format STT.ai supports — MP3, WAV, M4A, FLAC, OGG, MP4, MKV, MOV, WebM, AVI, and others. Output as TXT, SRT, VTT, DOCX, JSON, or PDF.
Yes. Speaker diarization runs alongside Whisper Turbo for every transcription — each speaker is labeled and you can rename them in the editor afterwards.
Yes. Whisper Turbo runs in our managed environment — audio is processed and deleted by default and never used for training without explicit opt-in. Pro plans add client-side encryption for transcripts at rest.
Use the compare-stt tool to run Whisper Turbo against any other supported model on the same audio — you'll see WER, segment count, speaker labels, and confidence scores side-by-side. The Whisper Turbo vs Whisper Large V3 comparison is the most commonly run.
Yes. Specify "whisper-turbo" as the model parameter on the /v1/transcribe endpoint. Python and Node.js SDKs include Whisper Turbo examples. Free API tier includes 100 minutes/month.
Yes. Because Whisper Turbo is MIT-licensed, you can self-host it. STT.ai's open-source page lists the project repo and weights. Most production teams use our hosted version to skip GPU procurement, model swaps, and ops.