Open ASR Leaderboard
An open-source automatic speech recognition (ASR) leaderboard hosted on Hugging Face that evaluates models on speech-to-text tasks using metrics such as word error rate.
Timeline (last 90 days)
Nemotron 3.5 Transcribe achieves an average word error rate (WER) of 4.71% on the English Open ASR Leaderboard.
Granite Speech 5.0 2B ranks second on the Open ASR Leaderboard with an average English word error rate (WER) of 5.68%.
The Open ASR Leaderboard is maintained by the Hugging Face community and mainly computes an average WER uniformly across multiple English datasets, while also providing reference metrics such as inference speed.
GPQA Diamond, MMLU-Pro, Humanity's Last Exam, SWE-bench Verified, Terminal-Bench, MMMU-Pro, LEXam, and Open ASR Leaderboard are all official benchmarks marked by Hugging Face as official benchmark.
All Facts (4)
Nemotron 3.5 Transcribe achieves an average word error rate (WER) of 4.71% on the English Open ASR Leaderboard.
50%UnverifiedGranite Speech 5.0 2B ranks second on the Open ASR Leaderboard with an average English word error rate (WER) of 5.68%.
50%UnverifiedThe Open ASR Leaderboard is maintained by the Hugging Face community and mainly computes an average WER uniformly across multiple English datasets, while also providing reference metrics such as inference speed.
50%UnverifiedGPQA Diamond, MMLU-Pro, Humanity's Last Exam, SWE-bench Verified, Terminal-Bench, MMMU-Pro, LEXam, and Open ASR Leaderboard are all official benchmarks marked by Hugging Face as official benchmark.
50%