You record a meeting, and an hour later you get a transcript riddled with mistakes in terminology and missing phrases during pauses. Sound familiar? We faced this regularly until we migrated all ASR integration pipelines to Whisper Large v3 — and WER dropped by half on complex audio.
Whisper Large v3 is OpenAI's flagship speech recognition model, supporting 99 languages. Compared to Large v2, it produces 10–20% fewer errors on most languages, including Russian. On clean Russian audio: 6–9% WER, on telephony: 15–20% WER. The model almost never hallucinates on silence and noise, handles punctuation better, and correctly manages code-switching (mixing languages in one dialogue). This is confirmed by independent benchmarks: according to OpenAI documentation, Whisper Large v3 tops multilingual benchmarks.
Why switch to Whisper Large v3?
Our migration experience from v2 showed: savings on transcript post-editing outweigh the implementation costs. We guarantee at least 10% WER reduction on your data — proven across dozens of projects. Self-hosted faster-whisper with int8 quantization runs up to 2x faster than the original implementation.
Comparison table (WER on Russian):
| Parameter | Large v2 | Large v3 |
|---|---|---|
| Clean | 8–12% | 6–9% |
| Telephony | 18–25% | 15–20% |
| Hallucinations on silence | Frequent | Rare |
| Punctuation | Average | Good |
| Code-switching | Weak | Good |
Setting up faster-whisper for production
For real-time you need a GPU with ≥10 GB VRAM. Optimal choices: NVIDIA A10G or RTX 4090. On CPU the model works but at 0.1–0.3× real-time — only for offline tasks.
Using faster-whisper with int8 quantization, the model fits in 6–7 GB VRAM at 1.5–2× real-time speed:
pip install faster-whisper from faster_whisper import WhisperModel model = WhisperModel( "large-v3", device="cuda", compute_type="int8_float16" ) segments, info = model.transcribe( "meeting.wav", language="ru", vad_filter=True, vad_parameters={"min_silence_duration_ms": 500} ) VAD filter is mandatory — it cuts out silence and noise, further reducing WER by 2–3% (30% fewer errors). The parameter min_silence_duration_ms adjusts sensitivity: 500 ms is a good balance for conversations.
Choosing between API and self-hosted
| Criterion | OpenAI API | Self-hosted (faster-whisper) |
|---|---|---|
| Speed of deployment | 1 day | 3–5 days |
| Data control | None | Full |
| Cost at high volumes | Increases | Fixed (hardware) |
| Streaming latency | Network | Minimal |
| Russian WER | 6–9% | 6–9% (with VAD) |
Self-hosted Whisper is advantageous if you process >100 hours of audio per month and value confidentiality. API is simpler for start and small volumes.
Use cases
- Audio transcription of meetings and interviews
- Automatic video subtitles
- Archival processing of call center audio databases for large-scale audio processing
For streaming transcription (e.g., live broadcast) we use int8 quantization with segment buffering — latency does not exceed 2–3 seconds.
Integration process
- Analytics: measure your audio, compute WER on a representative sample.
- Design: choose mode (API or self-hosted), pick hardware.
- Implementation: deploy the model, configure VAD, write conversion scripts.
- Testing: run on real data, record WER and speed.
- Deployment: launch to production, document, hand over to support.
Timeline: from 1 day (API) to 5 days (self-hosted with optimization). Cost is calculated individually based on audio volume and integration complexity. Self-hosted deployment costs typically range from $2,000 to $5,000, including optimization and documentation. Our clients save an average of 30% on post-editing costs.
Typical implementation mistakes
- Missing VAD leads to 10–15% extra errors. VAD is mandatory.
- Using CPU instead of GPU makes the model unsuitable for real-time.
- Skipping quantization wastes VRAM and slows inference.
- Incorrect batch_size (too large) causes OOM.
We've encountered these on nearly every second project and now bake the right settings in from the start.
What's included in the work
- Ready transcription pipeline (source code + configs)
- Optimal mode selection: API or self-hosted
- Whisper GPU optimization (quantization, batching)
- Operations and API documentation
- Team training (1–2 hours)
- Go-live support (2 weeks)
- Our MLOps ASR pipeline ensures smooth deployment and monitoring.
Our team has 5+ years of experience in ASR technology and has delivered 30+ projects. Each project is unique, but the approach is proven. Want to test Whisper Large v3 on your audio? Contact us — we'll send you a WER report with recommendations within 2 days. Get a consultation with an engineer right now.