SuperCLUE-ASR
Models transcribe Chinese speech under different acoustic conditions.
SuperCLUE reports scores in a public workbook.
- Made by
- SuperCLUE
- Kind
- Speech recognition and translation
- Level
- Everyday
- Scored
- Checked automatically
- Answers
- Short answer
- Measures
- Capability
- Language
- Chinese
- Interaction
- Single turn
- Refresh
- One fixed set
- Run by
- A host
- In views
- Audio and speech
- Last model added
- August 2026
- Scores refreshed
- November 20th, 2026
Hy-ASR-3.0-Preview tops the leaderboard with 6.65% GPT-Transcribe leads with 12.1% among models released by August 2026
Best score per entry from the benchmark's own leaderboard, read on November 20th, 2026. Where a model ran with several settings or harnesses, we keep its best. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders. The best model we track is GPT-Transcribe, at 12.1%. This view shows only models we track, each at its best.
The newest model with a score here came out in August 2026, and nobody has tested a newer one. Among the 20 highest-ranked models on the capability index, these have no score yet:
- GPT-6.2 Astra
- Claude Fable 5.2
-
Gemini 4 Carbon
- GPT-6 Astra
- and 16 more
- 1 Hy-ASR-3.0-Preview API 6.65%
- 2 StepAudio-3-ASR API 7.07%
- 3 fun-asr-realtime-2026-02-28 API 7.62%
- 4 豆包 SeedASR API 8.09%
- 5 讯飞录音文件转写大模型 API 8.88%
- 6 MAI-Transcribe-2 API 8.91%
- 7 Scribe-v2 API 9.15%
- 8 GPT-Transcribe API 12.1%
- 9
Gemini 3.5 Transcribe API 12.4%
- 10 MiMo-V2.5-ASR API 12.8%
- 11 GLM-ASR-2512 API 13%
- 12
Voxtral Mini Transcribe 2 API 15.7%
Source: Official SuperCLUE-ASR leaderboard
- 1 GPT-Transcribe API 12.1%
- 2
Gemini 3.5 Transcribe API 12.4%
- 3
Voxtral Mini Transcribe 2 API 15.7%
Source: Official SuperCLUE-ASR leaderboard
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Hy-ASR-3.0-Preview closed weights | 6.65% | API | Official leaderboard | |
| 2 | StepAudio-3-ASR closed weights | 7.07% | API | Official leaderboard | |
| 3 | fun-asr-realtime-2026-02-28 closed weights | 7.62% | API | Official leaderboard | |
| 4 | 豆包 SeedASR closed weights | 8.09% | API | Official leaderboard | |
| 5 | 讯飞录音文件转写大模型 closed weights | 8.88% | API | Official leaderboard | |
| 6 | MAI-Transcribe-2 closed weights | 8.91% | API | Official leaderboard | |
| 7 | Scribe-v2 closed weights | 9.15% | API | Official leaderboard | |
| 8 | GPT-Transcribe | 12.1% | API | Jul 2026 | Official leaderboard |
| 9 | | 12.4% | API | Aug 2026 | Official leaderboard |
| 10 | MiMo-V2.5-ASR open weights | 12.8% | API | Official leaderboard | |
| 11 | GLM-ASR-2512 closed weights | 13% | API | Official leaderboard | |
| 12 | | 15.7% | API | Feb 2026 | Official leaderboard |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | GPT-Transcribe | 12.1% | API | Jul 2026 | Official leaderboard |
| 2 | | 12.4% | API | Aug 2026 | Official leaderboard |
| 3 | | 15.7% | API | Feb 2026 | Official leaderboard |
Rankings by part
We read 23 parts from the benchmark’s own leaderboard in 5 groups: 基础中文转写, 真实声学鲁棒性, 对话与长音频, 专业领域与长尾信息 and 中文高难信息. Each ranks models on one part of the test. Gemini 3.5 Transcribe leads 12 of them, and 2 other models lead the rest. Arrows show how far a model moved from its overall rank among the models we track.
基础中文转写↓
3 models we track have a score here.
- 1 GPT-Transcribe 1.38%
- 2
Gemini 3.5 Transcribe 1.56%
- 3
Voxtral Mini Transcribe 2 1.70%
真实声学鲁棒性↓
3 models we track have a score here.
- 1 GPT-Transcribe 1.37%
- 2
Voxtral Mini Transcribe 2 ↑1 2.14%
- 3
Gemini 3.5 Transcribe ↓1 2.34%
对话与长音频↓
3 models we track have a score here.
- 1
Gemini 3.5 Transcribe ↑1 9.97%
- 2
Voxtral Mini Transcribe 2 ↑1 10.7%
- 3 GPT-Transcribe ↓2 10.8%
专业领域与长尾信息↓
3 models we track have a score here.
- 1
Gemini 3.5 Transcribe ↑1 6.53%
- 2 GPT-Transcribe ↓1 7.32%
- 3
Voxtral Mini Transcribe 2 10.2%
中文高难信息↓
3 models we track have a score here.
- 1 GPT-Transcribe 18.7%
- 2
Gemini 3.5 Transcribe 20.2%
- 3
Voxtral Mini Transcribe 2 25.7%
标准普通话↓
3 models we track have a score here.
- 1 GPT-Transcribe 0.19%
- 2
Gemini 3.5 Transcribe 0.91%
- 3
Voxtral Mini Transcribe 2 1.25%
自然口语普通话↓
3 models we track have a score here.
- 1
Voxtral Mini Transcribe 2 ↑2 2.15%
- 2
Gemini 3.5 Transcribe 2.21%
- 3 GPT-Transcribe ↓2 2.56%
环境噪声↓
3 models we track have a score here.
- 1
Gemini 3.5 Transcribe ↑1 0.55%
- 2 GPT-Transcribe ↓1 0.74%
- 3
Voxtral Mini Transcribe 2 1.00%
房间混响与远场↓
3 models we track have a score here.
- 1
Gemini 3.5 Transcribe ↑1 0.57%
- 2 GPT-Transcribe ↓1 0.85%
- 3
Voxtral Mini Transcribe 2 1.55%
通信与压缩↓
3 models we track have a score here.
- 1 GPT-Transcribe 3.61%
- 2
Voxtral Mini Transcribe 2 ↑1 3.91%
- 3
Gemini 3.5 Transcribe ↓1 6.03%
语速变化↓
3 models we track have a score here.
- 1 GPT-Transcribe 0.27%
- 2
Voxtral Mini Transcribe 2 ↑1 2.10%
- 3
Gemini 3.5 Transcribe ↓1 2.22%
双人自然对话↓
3 models we track have a score here.
- 1
Voxtral Mini Transcribe 2 ↑2 10.6%
- 2 GPT-Transcribe ↓1 10.6%
- 3
Gemini 3.5 Transcribe ↓1 11.2%
多人对话↓
3 models we track have a score here.
- 1
Gemini 3.5 Transcribe ↑1 11.2%
- 2 GPT-Transcribe ↓1 11.2%
- 3
Voxtral Mini Transcribe 2 12%
长音频连续转写↓
3 models we track have a score here.
- 1
Gemini 3.5 Transcribe ↑1 7.56%
- 2
Voxtral Mini Transcribe 2 ↑1 9.61%
- 3 GPT-Transcribe ↓2 10.5%
法律↓
3 models we track have a score here.
- 1
Gemini 3.5 Transcribe ↑1 12.2%
- 2 GPT-Transcribe ↓1 12.6%
- 3
Voxtral Mini Transcribe 2 17%
金融↓
3 models we track have a score here.
- 1
Gemini 3.5 Transcribe ↑1 3.01%
- 2 GPT-Transcribe ↓1 5.09%
- 3
Voxtral Mini Transcribe 2 7.55%
工程↓
3 models we track have a score here.
- 1 GPT-Transcribe 2.65%
- 2
Gemini 3.5 Transcribe 2.83%
- 3
Voxtral Mini Transcribe 2 4.89%
人工智能与信息技术↓
3 models we track have a score here.
- 1
Gemini 3.5 Transcribe ↑1 3.65%
- 2
Voxtral Mini Transcribe 2 ↑1 3.91%
- 3 GPT-Transcribe ↓2 5.09%
医疗↓
3 models we track have a score here.
- 1
Gemini 3.5 Transcribe ↑1 11%
- 2 GPT-Transcribe ↓1 11.2%
- 3
Voxtral Mini Transcribe 2 17.6%
数字相似↓
3 models we track have a score here.
- 1
Gemini 3.5 Transcribe ↑1 2.80%
- 2
Voxtral Mini Transcribe 2 ↑1 8.39%
- 3 GPT-Transcribe ↓2 10.7%
同音近音↓
3 models we track have a score here.
- 1 GPT-Transcribe 22%
- 2
Gemini 3.5 Transcribe 29.2%
- 3
Voxtral Mini Transcribe 2 29.9%
古文↓
3 models we track have a score here.
- 1 GPT-Transcribe 21.1%
- 2
Gemini 3.5 Transcribe 29.9%
- 3
Voxtral Mini Transcribe 2 39.6%
多音字↓
3 models we track have a score here.
- 1
Gemini 3.5 Transcribe ↑1 19%
- 2 GPT-Transcribe ↓1 21.1%
- 3
Voxtral Mini Transcribe 2 25.1%
2 models have held number one on the official leaderboard since 2023
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: Official SuperCLUE-ASR leaderboard
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Official SuperCLUE-ASR leaderboard
No lab model card reports it yet
We read the headline results of 540 launch posts from 57 labs, and none of them list this benchmark.
More in Audio and video
- Video-MMMU In use 300 expert-level lecture videos with questions testing perception, comprehension and adaptation of knowledge. Accuracy.
Qwen 3.8 Max 88.7%
- LVBench In use Hours-long video understanding with multiple-choice questions on reasoning, grounding and summarization.
Gemini 4 Argon 91.7%
- MMVU In use 3,000 expert-annotated questions on videos across 27 disciplines. Multiple-choice and open-ended accuracy.
Qwen 3.8 Max 82.4%
- MLVU In use Multi-task long video understanding (minutes to hours) with multiple-choice and generation tasks. Accuracy.
Qwen 3.8 Max 90.8%
- FLEURS In use Models transcribe read speech in 102 languages built from FLoRes-101 sentences. Model cards usually report word error rate, so lower scores are better.
Phi-4-multimodal 4.00%
- VoiceBench In use Spoken instructions with speaker, environment and content variation. Voice assistant responses scored by GPT judge or rules.
Qwen3.5-Omni-Plus 93.1%
Data audit
8 of 9 checks pass, 1 warning. Look first at the Task access check. We don’t know whether the tasks are public. Every benchmark
Catalog record 1 warning 1 pass
- Warning Task access We don’t know whether the tasks are public.
- Pass Confidence Checked against a primary source.
Scores 3 pass 2 notes
- Pass Independent runs Scores from the official leaderboard (12 entries).
- Pass Models scored 3 models we track, plus 9 other entries on its leaderboard.
- Pass Newest score The newest is from August 2026, for Gemini 3.5 Transcribe.
- Note Frontier models None of its models are on the capabilities index.
- Note Headroom Lower is better, so there’s no ceiling to reach.
Leaderboard import 3 pass
- Pass Importer We read the board on superclueai.com: 3 rows for models we track and 9 other entries at the last read.
- Pass Last read November 20th, 2026.
- Pass Read worked The last read worked.
Consistency 1 pass 1 note
- Note Lab claims No model has both a lab number and an independent run.
- Pass Known issues No published audit found problems.