Whisper Large v3
OpenAIAudio & Speech
92.0
Performance
★ 4.7
Rating
1560
Reviews
Speech RecognitionOpen SourceLarge
About
OpenAI's industry-standard open-source speech recognition model supporting 100+ languages with strong accuracy.
Strengths
Industry-standard for speech-to-text with support for 100+ languages. Strong accuracy across diverse accents, noise conditions, and audio quality. Open-source (MIT) so can run locally for free. Timestamp generation for subtitle creation.
Specifications
- Context window
- —
- Parameters
- 1.55B
Available On
OpenAI APIHuggingFaceReplicateLocal deploymentGroq
Features
speech to texttranslationtimestamp generationlanguage detectionmultilingual
Market Sentiment
Polymarket prediction market odds for "Best AI Model" — showing how the market values each provider over time.
Loading sentiment data...
Performance Trend
Benchmark score trends over time for the top 5 benchmarks.
Loading history...
Benchmarks
Scores from various benchmark tests; higher is better.
| Test | Score | Percentile | Source |
|---|---|---|---|
| WER (inverted) | 94.5 | p97 | seed |