Model card
Whisper Large v3.
OpenAIopen-source1.5B paramsTransformer encoder-decoder
OpenAI Whisper Large V3 multilingual ASR model. 1.5B parameters.
§ 02 · Benchmarks
Every benchmark Whisper Large v3 has a recorded score for.
| # | Benchmark | Area · Task | Metric | Value | Rank | Date | Source |
|---|---|---|---|---|---|---|---|
| 01 | LibriSpeech | Speech · Speech Recognition | wer-test-other | 3.6% | #3 | 2022-12-06 | source ↗ |
| 02 | Common Voice | Speech · Speech Recognition | wer | 8.4% | #4 | 2022-12-06 | source ↗ |
| 03 | LibriSpeech | Speech · Speech Recognition | wer-test-clean | 1.8% | #5 | 2022-12-06 | source ↗ |
| 04 | VoxPopuli | Speech · Speech Recognition | wer | 9.5% | #21 | — | source ↗ |
| 05 | TED-LIUM | Speech · Speech Recognition | wer | 3.9% | #34 | — | source ↗ |
| 06 | LibriSpeech | Speech · Speech Recognition | wer | 3.9% | #43 | — | source ↗ |
| 07 | Open ASR Leaderboard | Speech · Speech Recognition | rtfx | 145.51 | #43 | — | source ↗ |
| 08 | Open ASR Leaderboard | Speech · Speech Recognition | wer | 7.4% | #44 | — | source ↗ |
| 09 | SPGISpeech | Speech · Speech Recognition | wer | 2.9% | #44 | — | source ↗ |
Rank column shows this model’s position vs all other models scored on the same benchmark + metric (competitors after the slash). #1 in red means current SOTA. Sorted by rank, then newest result.
§ 04 · Papers
2 papers with results for Whisper Large v3.
- 2022-12-06· Speech· 3 results
Robust Speech Recognition via Large-Scale Weak Supervision (Whisper)
- 2022-12-06· 6 results
Robust Speech Recognition via Large-Scale Weak Supervision
§ 05 · Related models
Other OpenAI models scored on Codesota.
§ 06 · Sources & freshness
Where these numbers come from.
pwc-dump
6
results
arxiv
3
results
3 of 9 rows marked verified.