GigaSpeech (ESB test)
Multimodal · 2022-10-24
The GigaSpeech test split as packaged by the ESB benchmark and used by the Hugging Face Open ASR Leaderboard: 19,931 utterances of English audiobook, podcast and YouTube speech. Scores are corpus word error rate in percent after Whisper English normalization; lower is better.
Top models (lower is better)
| Model | Score |
|---|---|
| LFM2.5-Audio-1.5B | 10.5 |