Whisper Large V3
STT.ai runs this model on the paid tier. It is the model behind STT.ai Enhanced. Free transcription uses Whisper Large v3 Turbo, which is faster.
See plans →
·
Transcribe free with Whisper Turbo →
4.2%
WER
99
Languages
15.9x
Speed
MIT
License
About Whisper Large V3
Whisper Large V3 is OpenAI's flagship open-source speech recognition model. With 1.55 billion parameters, it offers exceptional accuracy across 99 languages. It uses a transformer encoder-decoder architecture trained on 680,000 hours of multilingual audio data.
Model Info
- ProviderOpenAI
- Architecture-
- LicenseMIT
- UpdatedAug 2026
Frequently Asked Questions
Whisper Large V3 is a speech-to-text model by OpenAI. STT.ai runs it on our own GPU for accounts on a paid plan, and it is the model behind the STT.ai Enhanced tier. Free transcription uses Whisper Large v3 Turbo, the faster checkpoint of the same family.
On standard benchmarks, Whisper Large V3 achieves around 4.2% 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 Large V3 runs on STT.ai's free tier — every visitor gets 600 minutes to start at no cost. Paid plans add longer per-file limits, private transcripts, and priority queueing.
Whisper Large V3 is released under MIT, a permissive open-source license. You can self-host Whisper Large V3 on your own hardware or use our hosted version — both are commercially usable.
Whisper Large V3 supports 99 languages. Auto-detection picks the right language for most audio; you can also specify it manually for a small accuracy lift.
Whisper Large V3 processes audio at about 15.9x real-time on our GPUs. A 1-hour audio file finishes in under 3 minutes; longer files queue and notify by email when done.
Whisper Large V3 has 1.55B parameters. Larger models tend to be more accurate but slower; STT.ai hosts Whisper Large V3 on GPU so the parameter count doesn't affect your client-side performance.
Whisper Large V3 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 Large V3 for every transcription — each speaker is labeled and you can rename them in the editor afterwards.
Yes. Whisper Large V3 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 Large V3 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 Large V3 vs Whisper Large V3 comparison is the most commonly run.
Yes, on a paid plan. Specify "whisper-large-v3" as the model parameter on the /v1/transcribe endpoint. Keys without an active plan are served Whisper Large v3 Turbo instead. The request still succeeds; it just runs the free model.
Yes. Because Whisper Large V3 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.