Automatic Speech Recognition on Common Voice 15
6.1WERQwen3-Omni (Sep 25)
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| Qwen3-Omni (Sep 25)Wer=06.12026.04 | 6.1 | — | — | — | — | — | |
| AFlamingo3 (NeurIPS 25)Model Category=Audio MLLMs2026.04 | 7.4 | — | — | — | — | — | |
| Audio Flamingo 3Backbone=Audio Flamingo 3, Method=Baseline2026.06 | 7.4 | — | — | — | — | — | |
| Audio Flamingo 3 + CoATBackbone=Audio Flamingo 3, Method=CoAT2026.06 | 7.4 | — | — | — | — | — | |
| Qwen2.5-Omni (Mar 25)Model Category=Omnimodal MLLMs2026.04 | 7.6 | — | — | — | — | — | |
| Qwen2.5-Omni-7BBackbone=Qwen2.5-Omni-7B, Method=Baseline2026.06 | 7.6 | — | — | — | — | — | |
| Qwen2.5-Omni-7B + CoATBackbone=Qwen2.5-Omni-7B, Method=CoAT2026.06 | 7.68 | — | — | — | — | — | |
| Omni-o3Model Category=Omnimodal MLLMs2026.04 | 7.9 | — | — | — | — | — | |
| Qwen2-Audio (Jul 24)Model Category=Audio MLLMs2026.04 | 8.6 | — | — | — | — | — | |
| Qwen2-Audio + CoATBackbone=Qwen2-Audio, Method=CoAT2026.06 | 9.25 | — | — | — | — | — | |
| AudioReaonser (Sep 25)Model Category=Audio MLLMs2026.04 | 9.5 | — | — | — | — | — | |
| Gemini 2.5 Pro (May 25)Model Category=Closed-Source APIs2026.04 | 9.9 | — | — | — | — | — | |
| GPT-4o (May 24)Model Category=Closed-Source APIs2026.04 | 10 | — | — | — | — | — | |
| Ola (Feb 25)Model Category=Omnimodal MLLMs2026.04 | 13.1 | — | — | — | — | — | |
| DeSTA 2.5 (Jul 25)Model Category=Audio MLLMs2026.04 | 20.1 | — | — | — | — | — | |
| Qwen2-AudioBackbone=Qwen2-Audio, Method=Baseline2026.06 | 21.48 | — | — | — | — | — | |
| FastSLMNumber of Parameters=4.8B2026.01 | — | 10.9 | 4.55 | — | — | — | |
| Gemini-2.5-Flash2026.01 | — | 11.2 | 6.09 | — | — | — | |
| Gemini-3.1 Pro2026.04 | — | — | — | 8.59 | 13.4 | 6.78 | |
| Phi4-MultimodalNumber of Parameters=5.8B2026.01 | — | 7.61 | — | — | — | — | |
| Qwen2-AudioNumber of Parameters=8B2026.01 | — | 8.68 | — | — | — | — | |
| Qwen3.5-Omni-Flash2026.04 | — | — | — | 4.25 | 3.45 | 2.68 | |
| Qwen3.5-Omni-Plus2026.04 | — | — | — | 3.46 | 1.95 | 2.27 | |
| Voxtral-miniNumber of Parameters=4.7B2026.01 | — | 10.2 | — | — | — | — | |
| WhisperNumber of Parameters=1.5B2026.01 | — | 9.3 | 5.74 | — | — | — |