AudioBench - IDPC Short WER: leaderboard
AudioBench IDPC Short word error rate slice measuring automatic speech-recognition transcription quality, where lower scores indicate fewer errors.
Metric: WER (%) (lower is better). Source: github.com. 8 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | cascade_whisper_large_v3_llama_3_8b_instruct | 15.8 | |
| 2 | whisper_large_v3 | 16.63 | |
| 3 | cascade_whisper_large_v2_gemma2_9b_cpt_sea_lionv3_instruct | 16.93 | |
| 4 | Qwen2-Audio-7B-Instruct | 21.33 | #880 |
| 5 | MERaLiON-AudioLLM-Whisper-SEA-LION | 24.92 | |
| 6 | SALMONN_7B | 26.31 | |
| 7 | WavLLM_fairseq | 36.73 | |
| 8 | Qwen-Audio-Chat | 60.08 | #1244 |
Interactive version: theaggregate.ai/benchmark?slug=audiobench-idpc-short-wer · How It Works · Data refreshed daily, snapshot 2026-10-20.