Uzbek Open ASR Leaderboard
Models transcribe Uzbek recordings from four test sets. The main score averages the four word error rates with equal weight.
NavAI scored 34 systems across architectures. One system trained on FLEURS test audio. Its overall score includes that overlap.
- Made by
- NavAI
- Released
- July 2026
- Kind
- Speech recognition and translation
- Level
- Everyday
- Scored
- Checked automatically
- Answers
- Open-ended
- Tasks
- All public
- Measures
- Capability
- Language
- Uzbek
- Interaction
- Single turn
- Items from
- Another benchmark
- Refresh
- One fixed set
- Run by
- Anyone
- In views
- Audio and speech
islomov/rubaistt_v2_medium tops the leaderboard with 8.57%
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.
- 1 islomov/rubaistt_v2_medium Whisper / FLEURS test data in training 8.57%
- 2 ai-sage/GigaAM-Multilingual (Large) Conformer / Single-pass decoding 9.61%
- 3 Kotib/uzbek_stt_v1 Whisper / Training data undisclosed 9.62%
- 4 navai-uz/whisper-medium-uzbek Whisper 9.73%
- 5 nvidia/stt_uz_fastconformer_hybrid_large_pc FastConformer / Single-pass decoding 9.93%
- 6 uzinfocom-edu-ai/asr-uz-fastconformer-large FastConformer / Single-pass decoding 10.1%
- 7 navai-uz/whisper-small-uzbek Whisper 11.6%
- 8 ai-sage/GigaAM-Multilingual (220M) Conformer / Single-pass decoding 11.6%
- 9 facebook/seamless-m4t-v2-large Seamless / Single-pass decoding 14.2%
- 10 OvozifyLabs/whisper-small-uz-v1 Whisper 14.3%
- 11 Abduqayum/whisper-uzbek-medium-callcenter Whisper 15.1%
- 12 navai-uz/whisper-base-uzbek Whisper 15.1%
- 13 mrmuminov/whisper-small-uz Whisper 15.3%
- 14 aisha-org/Whisper-Uzbek Whisper 17.5%
- 15 navai-uz/whisper-tiny-uzbek Whisper 18.1%
- 16 Yorkinjon/whisper-medium-uzbek-ynv1 Whisper 18.7%
- 17 Yorkinjon/whisper-small-uzbek-ynv3 Whisper 22.8%
- 18 abduaziz/whisper-small-uzbek Whisper 23.4%
- 19 Yorkinjon/whisper-small-uzbek-ynv2 Whisper 23.7%
- 20 blackhole33/whisper-small-uz Whisper 23.9%
- 21 mustafoyev202/whisper-uz Whisper 25.3%
- 22 Yorkinjon/whisper-small-uzbek-yn Whisper 25.6%
- 23 firdavsus/whisper-small-uz-ru Whisper 26.1%
- 24 oyqiz/uzbek_stt wav2vec2-CTC / Single-pass decoding 27.3%
- 25 jmshd/whisper-uz Whisper 28.5%
- 26 tukhtashevshohruh/whisper-small-uz Whisper 28.6%
- 27 azamat45/whisper-small-uz Whisper 33.2%
- 28 blackhole33/whisper-small-uzb-commonvoice Whisper 33.6%
- 29 Inomjonov/mironshoh-whisper-medium-uz-10k Whisper 34.5%
- 30 mirodil/whisper-small-uzbek Whisper 35%
- 31 mirodil/whisper-tiny-uzbek Whisper 35.3%
- 32 mrmuminov/whisper-tiny-uz Whisper 35.8%
- 33 Xojakbar/whisper-small-uz Whisper 50.3%
- 34 Saidakbar01/whisper-uz-finetuned Whisper 68.7%
Source: Official Uzbek Open ASR Leaderboard leaderboard
| # | Entry | Score | Setting | Submitted by | Date | Source |
|---|---|---|---|---|---|---|
| 1 | islomov/rubaistt_v2_medium | 8.57% | Whisper / FLEURS test data in training | Official leaderboard | ||
| 2 | ai-sage/GigaAM-Multilingual (Large) | 9.61% | Conformer / Single-pass decoding | Official leaderboard | ||
| 3 | Kotib/uzbek_stt_v1 | 9.62% | Whisper / Training data undisclosed | Official leaderboard | ||
| 4 | navai-uz/whisper-medium-uzbek | 9.73% | Whisper | NavAI | Official leaderboard | |
| 5 | nvidia/stt_uz_fastconformer_hybrid_large_pc | 9.93% | FastConformer / Single-pass decoding | Official leaderboard | ||
| 6 | uzinfocom-edu-ai/asr-uz-fastconformer-large | 10.1% | FastConformer / Single-pass decoding | Official leaderboard | ||
| 7 | navai-uz/whisper-small-uzbek | 11.6% | Whisper | NavAI | Official leaderboard | |
| 8 | ai-sage/GigaAM-Multilingual (220M) | 11.6% | Conformer / Single-pass decoding | Official leaderboard | ||
| 9 | facebook/seamless-m4t-v2-large | 14.2% | Seamless / Single-pass decoding | Official leaderboard | ||
| 10 | OvozifyLabs/whisper-small-uz-v1 | 14.3% | Whisper | Official leaderboard | ||
| 11 | Abduqayum/whisper-uzbek-medium-callcenter | 15.1% | Whisper | Official leaderboard | ||
| 12 | navai-uz/whisper-base-uzbek | 15.1% | Whisper | NavAI | Official leaderboard | |
| 13 | mrmuminov/whisper-small-uz | 15.3% | Whisper | Official leaderboard | ||
| 14 | aisha-org/Whisper-Uzbek | 17.5% | Whisper | Official leaderboard | ||
| 15 | navai-uz/whisper-tiny-uzbek | 18.1% | Whisper | NavAI | Official leaderboard | |
| 16 | Yorkinjon/whisper-medium-uzbek-ynv1 | 18.7% | Whisper | Official leaderboard | ||
| 17 | Yorkinjon/whisper-small-uzbek-ynv3 | 22.8% | Whisper | Official leaderboard | ||
| 18 | abduaziz/whisper-small-uzbek | 23.4% | Whisper | Official leaderboard | ||
| 19 | Yorkinjon/whisper-small-uzbek-ynv2 | 23.7% | Whisper | Official leaderboard | ||
| 20 | blackhole33/whisper-small-uz | 23.9% | Whisper | Official leaderboard | ||
| 21 | mustafoyev202/whisper-uz | 25.3% | Whisper | Official leaderboard | ||
| 22 | Yorkinjon/whisper-small-uzbek-yn | 25.6% | Whisper | Official leaderboard | ||
| 23 | firdavsus/whisper-small-uz-ru | 26.1% | Whisper | Official leaderboard | ||
| 24 | oyqiz/uzbek_stt | 27.3% | wav2vec2-CTC / Single-pass decoding | Official leaderboard | ||
| 25 | jmshd/whisper-uz | 28.5% | Whisper | Official leaderboard | ||
| 26 | tukhtashevshohruh/whisper-small-uz | 28.6% | Whisper | Official leaderboard | ||
| 27 | azamat45/whisper-small-uz | 33.2% | Whisper | Official leaderboard | ||
| 28 | blackhole33/whisper-small-uzb-commonvoice | 33.6% | Whisper | Official leaderboard | ||
| 29 | Inomjonov/mironshoh-whisper-medium-uz-10k | 34.5% | Whisper | Official leaderboard | ||
| 30 | mirodil/whisper-small-uzbek | 35% | Whisper | Official leaderboard | ||
| 31 | mirodil/whisper-tiny-uzbek | 35.3% | Whisper | Official leaderboard | ||
| 32 | mrmuminov/whisper-tiny-uz | 35.8% | Whisper | Official leaderboard | ||
| 33 | Xojakbar/whisper-small-uz | 50.3% | Whisper | Official leaderboard | ||
| 34 | Saidakbar01/whisper-uz-finetuned | 68.7% | Whisper | Official 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 8 checks pass. Every benchmark
Catalog record 2 pass
- Pass Task access All tasks are public.
- Pass Confidence Checked against a primary source.
Scores 2 pass 3 notes
- Pass Independent runs Scores from the official leaderboard (34 entries).
- Pass Models scored 34 entries on its leaderboard, none a model we track.
- Note Newest score No scores yet.
- Note Frontier models No models scored yet.
- Note Headroom Lower is better, so there’s no ceiling to reach.
Leaderboard import 3 pass
- Pass Importer We read the board on github.com: 34 entries at the last read, none a model we track.
- 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.