ContextASR-Bench
Models transcribe speech with optional domain labels and named entities. The benchmark measures transcription errors and recognition of named entities.
- Made by
- He Wang and colleagues
- Released
- July 2025
- Kind
- Speech recognition and translation
- Level
- Everyday
- Scored
- Checked automatically
- Answers
- Open-ended
- Tasks
- All public
- Measures
- Capability
- Interaction
- Single turn
- Refresh
- One fixed set
- Run by
- Anyone
- In views
- Audio and speech
- Last model added
- April 2025
- Scores refreshed
- November 20th, 2026
Kimi-Audio-7B-Instruct leads with 3.10% among models released by April 2025 Kimi-Audio-7B-Instruct leads with 3.10% among models released by April 2025
Best score per entry from the benchmark's paper, read on November 20th, 2026. A paper’s table doesn’t add newer models. The headline averages word error rate across the four language and speech-format combinations. Where a model ran with several settings or harnesses, we keep its best. Every run is in the next chart. Names in gray aren't in our model list: fine-tunes, agent systems, products and older models. They don't count toward leaders. This view shows only models we track, each at its best.
The newest model with a score here came out in April 2025, 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 Kimi-Audio-7B-Instruct Fine-grained context 3.10%
- 2
Qwen2.5-Omni-7B Fine-grained context 3.38%
- 3 Qwen2.5-Omni-3B Fine-grained context 3.52%
- 4
Baichuan-Audio-Instruct Fine-grained context 4.58%
- 5 FireredASR-LLM-L No context 4.88%
- 6
Baichuan-Omni-1.5 Fine-grained context 6.51%
- 7 FireredASR-AED-L No context 9.36%
- 8 Sensevoice-Small No context 9.88%
- 9 Whisper large-v3 No context 10.4%
- 10
Qwen2-Audio-7B-Instruct Fine-grained context 10.6%
- 11 Whisper large-v3-turbo No context 11%
- 12 Paraformer-Large No context 18.4%
Source: The ContextASR-Bench paper
- 1 Kimi-Audio-7B-Instruct Fine-grained context 3.10%
- 2
Qwen2.5-Omni-7B Fine-grained context 3.38%
- 3
Baichuan-Audio-Instruct Fine-grained context 4.58%
- 4
Baichuan-Omni-1.5 Fine-grained context 6.51%
- 5 Whisper large-v3 No context 10.4%
- 6
Qwen2-Audio-7B-Instruct Fine-grained context 10.6%
- 7 Whisper large-v3-turbo No context 11%
Source: The ContextASR-Bench paper
| # | Entry | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Kimi-Audio-7B-Instruct | 3.10% | Fine-grained context | Apr 2025 | Official leaderboard |
| 2 | | 3.38% | Fine-grained context | Mar 2025 | Official leaderboard |
| 3 | Qwen2.5-Omni-3B 5.4B | 3.52% | Fine-grained context | Official leaderboard | |
| 4 | | 4.58% | Fine-grained context | Feb 2025 | Official leaderboard |
| 5 | FireredASR-LLM-L 8.3B | 4.88% | No context | Official leaderboard | |
| 6 | | 6.51% | Fine-grained context | Jan 2025 | Official leaderboard |
| 7 | FireredASR-AED-L 1.1B | 9.36% | No context | Official leaderboard | |
| 8 | Sensevoice-Small 0.2B | 9.88% | No context | Official leaderboard | |
| 9 | Whisper large-v3 | 10.4% | No context | Nov 2023 | Official leaderboard |
| 10 | | 10.6% | Fine-grained context | Aug 2024 | Official leaderboard |
| 11 | Whisper large-v3-turbo | 11% | No context | Oct 2024 | Official leaderboard |
| 12 | Paraformer-Large 0.2B | 18.4% | No context | Official leaderboard |
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | Kimi-Audio-7B-Instruct | 3.10% | Fine-grained context | Apr 2025 | Official leaderboard |
| 2 | | 3.38% | Fine-grained context | Mar 2025 | Official leaderboard |
| 3 | | 4.58% | Fine-grained context | Feb 2025 | Official leaderboard |
| 4 | | 6.51% | Fine-grained context | Jan 2025 | Official leaderboard |
| 5 | Whisper large-v3 | 10.4% | No context | Nov 2023 | Official leaderboard |
| 6 | | 10.6% | Fine-grained context | Aug 2024 | Official leaderboard |
| 7 | Whisper large-v3-turbo | 11% | No context | Oct 2024 | Official leaderboard |
Rankings by part
We read 12 parts from the benchmark’s paper, each ranking models on one part of the test, such as Speech-EN: WER, Speech-EN: NE-WER and Speech-EN: NE-FNR. Baichuan-Audio-Instruct leads 8 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.
Speech-EN: WER
7 models we track have a score here.
- 1 Kimi-Audio-7B-Instruct 2.90%
- 2
Qwen2.5-Omni-7B 3.96%
- 3
Baichuan-Audio-Instruct 7.52%
- 4
Baichuan-Omni-1.5 8.16%
- 5 Whisper large-v3 9.36%
- 6 Whisper large-v3-turbo ↑1 9.84%
- 7
Qwen2-Audio-7B-Instruct ↓1 11.5%
Speech-EN: NE-WER
7 models we track have a score here.
- 1
Baichuan-Audio-Instruct ↑2 5.87%
- 2 Kimi-Audio-7B-Instruct ↓1 6.68%
- 3
Qwen2.5-Omni-7B ↓1 7.38%
- 4
Baichuan-Omni-1.5 7.69%
- 5
Qwen2-Audio-7B-Instruct ↑1 27.3%
- 6 Whisper large-v3 ↓1 29.6%
- 7 Whisper large-v3-turbo 32.1%
Speech-EN: NE-FNR
7 models we track have a score here.
- 1
Baichuan-Audio-Instruct ↑2 4.55%
- 2
Baichuan-Omni-1.5 ↑2 6.53%
- 3 Kimi-Audio-7B-Instruct ↓2 8.01%
- 4
Qwen2.5-Omni-7B ↓2 8.72%
- 5
Qwen2-Audio-7B-Instruct ↑1 35.1%
- 6 Whisper large-v3 ↓1 39.9%
- 7 Whisper large-v3-turbo 44%
Dialogue-EN: WER
7 models we track have a score here.
- 1 Kimi-Audio-7B-Instruct 4.58%
- 2
Qwen2.5-Omni-7B 5.32%
- 3
Baichuan-Audio-Instruct 5.66%
- 4 Whisper large-v3-turbo ↑3 9.36%
- 5 Whisper large-v3 9.62%
- 6
Baichuan-Omni-1.5 ↓2 9.86%
- 7
Qwen2-Audio-7B-Instruct ↓1 13.8%
Dialogue-EN: NE-WER
7 models we track have a score here.
- 1
Baichuan-Audio-Instruct ↑2 10%
- 2
Qwen2.5-Omni-7B 11.8%
- 3 Kimi-Audio-7B-Instruct ↓2 13.5%
- 4
Baichuan-Omni-1.5 14.4%
- 5
Qwen2-Audio-7B-Instruct ↑1 33%
- 6 Whisper large-v3 ↓1 33.5%
- 7 Whisper large-v3-turbo 34.7%
Dialogue-EN: NE-FNR
7 models we track have a score here.
- 1
Baichuan-Audio-Instruct ↑2 3.64%
- 2
Baichuan-Omni-1.5 ↑2 5.54%
- 3
Qwen2.5-Omni-7B ↓1 9.24%
- 4 Kimi-Audio-7B-Instruct ↓3 11.3%
- 5
Qwen2-Audio-7B-Instruct ↑1 32.9%
- 6 Whisper large-v3 ↓1 35.2%
- 7 Whisper large-v3-turbo 36.7%
Speech-ZH: WER
7 models we track have a score here.
- 1
Qwen2.5-Omni-7B ↑1 1.84%
- 2 Kimi-Audio-7B-Instruct ↓1 1.95%
- 3
Baichuan-Audio-Instruct 2.16%
- 4
Baichuan-Omni-1.5 2.98%
- 5
Qwen2-Audio-7B-Instruct ↑1 9.92%
- 6 Whisper large-v3 ↓1 13.6%
- 7 Whisper large-v3-turbo 14.7%
Speech-ZH: NE-WER
7 models we track have a score here.
- 1
Baichuan-Audio-Instruct ↑2 6.65%
- 2
Baichuan-Omni-1.5 ↑2 8.39%
- 3
Qwen2.5-Omni-7B ↓1 9.80%
- 4 Kimi-Audio-7B-Instruct ↓3 11.1%
- 5
Qwen2-Audio-7B-Instruct ↑1 24.1%
- 6 Whisper large-v3 ↓1 46.6%
- 7 Whisper large-v3-turbo 49.5%
Speech-ZH: NE-FNR
7 models we track have a score here.
- 1
Baichuan-Audio-Instruct ↑2 2.35%
- 2
Baichuan-Omni-1.5 ↑2 4.71%
- 3
Qwen2.5-Omni-7B ↓1 12.2%
- 4 Kimi-Audio-7B-Instruct ↓3 15.3%
- 5
Qwen2-Audio-7B-Instruct ↑1 30%
- 6 Whisper large-v3 ↓1 77.3%
- 7 Whisper large-v3-turbo 82.2%
Dialogue-ZH: WER
7 models we track have a score here.
- 1
Qwen2.5-Omni-7B ↑1 2.40%
- 2 Kimi-Audio-7B-Instruct ↓1 2.90%
- 3
Baichuan-Audio-Instruct 2.96%
- 4
Baichuan-Omni-1.5 5.00%
- 5
Qwen2-Audio-7B-Instruct ↑1 7.00%
- 6 Whisper large-v3 ↓1 9.05%
- 7 Whisper large-v3-turbo 10.1%
Dialogue-ZH: NE-WER
7 models we track have a score here.
- 1
Baichuan-Audio-Instruct ↑2 11.5%
- 2
Qwen2.5-Omni-7B 14.1%
- 3 Kimi-Audio-7B-Instruct ↓2 15.9%
- 4
Baichuan-Omni-1.5 16.8%
- 5
Qwen2-Audio-7B-Instruct ↑1 22.8%
- 6 Whisper large-v3 ↓1 44.8%
- 7 Whisper large-v3-turbo 47.2%
Dialogue-ZH: NE-FNR
7 models we track have a score here.
- 1
Baichuan-Audio-Instruct ↑2 3.94%
- 2
Baichuan-Omni-1.5 ↑2 7.84%
- 3
Qwen2.5-Omni-7B ↓1 13.2%
- 4 Kimi-Audio-7B-Instruct ↓3 16.7%
- 5
Qwen2-Audio-7B-Instruct ↑1 26.2%
- 6 Whisper large-v3 ↓1 62.3%
- 7 Whisper large-v3-turbo 66.6%
Scores at each effort level
Models the leaderboard ran at two or more effort levels. We don’t have enough cost data here to say what the extra effort costs.
Source: The ContextASR-Bench paper
| Model | Effort | Harness | Score | Cost per task | Output tokens |
|---|---|---|---|---|---|
| No context | 11.3% | ||||
| Coarse-grained context | 11.3% | ||||
| Whisper large-v3-turbo | No context | 11% | |||
| Fine-grained context | 10.6% | ||||
| Whisper large-v3 | No context | 10.4% | |||
| No context | 8.90% | ||||
| No context | 7.63% | ||||
| Coarse-grained context | 7.47% | ||||
| Fine-grained context | 6.51% | ||||
| Coarse-grained context | 5.67% | ||||
| Fine-grained context | 4.58% | ||||
| Coarse-grained context | 4.55% | ||||
| No context | 4.42% | ||||
| Kimi-Audio-7B-Instruct | Coarse-grained context | 3.77% | |||
| Kimi-Audio-7B-Instruct | No context | 3.68% | |||
| Fine-grained context | 3.38% | ||||
| Kimi-Audio-7B-Instruct | Fine-grained context | 3.10% |
Scores at each effort level
Models the leaderboard ran at two or more effort levels. We don’t have enough cost data here to say what the extra effort costs.
Source: The ContextASR-Bench paper
| Model | Effort | Harness | Score | Cost per task | Output tokens |
|---|---|---|---|---|---|
| No context | 11.3% | ||||
| Coarse-grained context | 11.3% | ||||
| Whisper large-v3-turbo | No context | 11% | |||
| Fine-grained context | 10.6% | ||||
| Whisper large-v3 | No context | 10.4% | |||
| No context | 8.90% | ||||
| No context | 7.63% | ||||
| Coarse-grained context | 7.47% | ||||
| Fine-grained context | 6.51% | ||||
| Coarse-grained context | 5.67% | ||||
| Fine-grained context | 4.58% | ||||
| Coarse-grained context | 4.55% | ||||
| No context | 4.42% | ||||
| Kimi-Audio-7B-Instruct | Coarse-grained context | 3.77% | |||
| Kimi-Audio-7B-Instruct | No context | 3.68% | |||
| Fine-grained context | 3.38% | ||||
| Kimi-Audio-7B-Instruct | Fine-grained context | 3.10% |
Number one over time
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: The ContextASR-Bench paper
Scores over time
All the scores we have, one dot each, so a model tested at several settings appears more than once.
Source: Official ContextASR-Bench 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 Newest score check. The newest came 19 months ago, in April 2025, for Kimi-Audio-7B-Instruct. Every benchmark
Catalog record 2 pass
- Pass Task access All tasks are public.
- Pass Confidence Checked against a primary source.
Scores 1 warning 2 pass 2 notes
- Pass Independent runs Scores from the benchmark’s paper (12 entries).
- Pass Models scored 7 models we track, plus 5 other entries on its leaderboard.
- Warning Newest score The newest came 19 months ago, in April 2025, for Kimi-Audio-7B-Instruct.
- 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 arxiv.org: 17 rows for models we track and 5 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.