Garble-Bench (Voice Arena)
Models respond to dialogue with speech recognition errors. The evaluation measures error detection, transcription fidelity, conversation recovery and task completion.
Voice Arena published results for eight model configurations and the scripts that score them. Its unrelated namesake tests typing errors.
- Made by
- Voice Arena and collaborators
- Released
- September 2026
- Kind
- Audio understanding and voice
- Level
- Everyday
- Scored
- Graded by a model
- Answers
- Open-ended
- Tasks
- All public
- Measures
- Capability
- Language
- English
- Interaction
- Multi-turn
- Items from
- Generated
- Refresh
- One fixed set
- Run by
- Anyone
- Strongest model tested
-
Gemini 3.8 Flash #24 on the capability index
- Last model added
- September 2026
- Scores refreshed
- November 21st, 2026
Gemini 3.8 Flash leads with 0.98 among models released by September 2026
Best score per model from the benchmark's own leaderboard, read on November 21st, 2026. The source reports no overall score, so this is the plain mean of transactional and conversational task completion. Where a model ran with several settings or harnesses, we keep its best.
The newest model with a score here came out in September 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-6.1 Sol
- and 16 more
- 1
Gemini 3.8 Flash 0.98
- 2
Gemini 3.1 Pro 0.96
- 3 GPT-6 Luna 0.93
- 4 GLM-4.7 0.81
- 5 Mercury 2.5 0.80
- 6
Qwen3-32B 0.71
- 7
Qwen3-30B-A3B 0.63
Source: Official Garble-Bench (Voice Arena) leaderboard
| # | Model | Score | Setting | Date | Source |
|---|---|---|---|---|---|
| 1 | | 0.98 | Sep 2026 | Official leaderboard | |
| 2 | | 0.96 | Feb 2026 | Official leaderboard | |
| 3 | GPT-6 Luna | 0.93 | Sep 2026 | Official leaderboard | |
| 4 | GLM-4.7 | 0.81 | Dec 2025 | Official leaderboard | |
| 5 | Mercury 2.5 | 0.80 | Sep 2026 | Official leaderboard | |
| 6 | | 0.71 | Apr 2025 | Official leaderboard | |
| 7 | | 0.63 | Apr 2025 | Official leaderboard |
2 models have held number one on the official leaderboard since its release in September 2026
Lab claims count from the day the card came out, independent runs from the model’s release.
Source: Official Garble-Bench (Voice Arena) 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 Garble-Bench (Voice Arena) 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
All 9 checks pass. Every benchmark
Catalog record 2 pass
- Pass Task access All tasks are public.
- Pass Confidence Checked against a primary source.
Scores 3 pass 2 notes
- Pass Independent runs Scores from the official leaderboard (7 entries).
- Pass Models scored 7 models we track.
- Pass Newest score The newest is from September 2026, for GPT-6 Luna.
- Note Frontier models None of the top 10 has a score. The strongest tested is Gemini 3.8 Flash (#24).
- Note Headroom Scored as a plain number, with no fixed ceiling.
Leaderboard import 3 pass
- Pass Importer We read the board on github.com: 8 rows for models we track at the last read.
- Pass Last read November 21st, 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.