Skip to content

The Aggregate

Unified LLM rankings, daily

Models What's New Trends Benchmarks Skill Maps
How It Works Guesswork Metrics
The Aggregate
Loading data...

AudioBench - Common Voice 15 EN WER — leaderboard

Metric: WER (%) (lower is better). Source: github.com. 10 models tracked.

Top models

#ModelScore
1Phi-4 Multimodal Instruct8.26
2Qwen2-Audio-7B-Instruct11.44
3SALMONN-7B30.62

Interactive version: theaggregate.ai/benchmark?slug=audiobench-common-voice-15-en-wer · How the rankings work · Data refreshed daily, snapshot 2026-07-22.

Built by Mikhail Doroshenko — AI researcher, co-author of Humanity’s Last Exam. Independent project; no affiliation with any AI lab.