FeruzaSpeech
Models transcribe one woman reading Uzbek. The reference text is available in Latin and Cyrillic scripts.
NavAI scores 34 systems on the FeruzaSpeech test split. The authors require approval for access to recordings for academic research.
- Made by
- Anna Povey and Katherine Povey
- Released
- March 2024
- Kind
- Speech recognition and translation
- Level
- Everyday
- Scored
- Checked automatically
- Answers
- Open-ended
- Tasks
- Kept private
- Measures
- Capability
- Language
- Uzbek
- Interaction
- Single turn
- Items from
- Written by crowdworkers
- Refresh
- One fixed set
- Run by
- Anyone
- In views
- Audio and speech
Kotib/uzbek_stt_v1 tops the leaderboard with 6.86%
Best score per entry from NavAI’s 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 Kotib/uzbek_stt_v1 Whisper / Training data undisclosed 6.86%
- 2 nvidia/stt_uz_fastconformer_hybrid_large_pc FastConformer / Single-pass decoding 7.01%
- 3 uzinfocom-edu-ai/asr-uz-fastconformer-large FastConformer / Single-pass decoding 7.06%
- 4 navai-uz/whisper-medium-uzbek Whisper 7.42%
- 5 islomov/rubaistt_v2_medium Whisper / FLEURS test data in training 7.64%
- 6 navai-uz/whisper-small-uzbek Whisper 7.92%
- 7 OvozifyLabs/whisper-small-uz-v1 Whisper 8.07%
- 8 Abduqayum/whisper-uzbek-medium-callcenter Whisper 8.28%
- 9 navai-uz/whisper-base-uzbek Whisper 10.3%
- 10 aisha-org/Whisper-Uzbek Whisper 10.6%
- 11 Yorkinjon/whisper-medium-uzbek-ynv1 Whisper 11.4%
- 12 navai-uz/whisper-tiny-uzbek Whisper 11.5%
- 13 facebook/seamless-m4t-v2-large Seamless / Single-pass decoding 11.8%
- 14 mrmuminov/whisper-small-uz Whisper 12%
- 15 ai-sage/GigaAM-Multilingual (Large) Conformer / Single-pass decoding 12.9%
- 16 ai-sage/GigaAM-Multilingual (220M) Conformer / Single-pass decoding 13.4%
- 17 Yorkinjon/whisper-small-uzbek-ynv3 Whisper 13.9%
- 18 blackhole33/whisper-small-uz Whisper 14.2%
- 19 abduaziz/whisper-small-uzbek Whisper 14.2%
- 20 Yorkinjon/whisper-small-uzbek-ynv2 Whisper 14.4%
- 21 firdavsus/whisper-small-uz-ru Whisper 14.8%
- 22 mustafoyev202/whisper-uz Whisper 14.8%
- 23 Yorkinjon/whisper-small-uzbek-yn Whisper 15.6%
- 24 jmshd/whisper-uz Whisper 16.2%
- 25 tukhtashevshohruh/whisper-small-uz Whisper 17.4%
- 26 oyqiz/uzbek_stt wav2vec2-CTC / Single-pass decoding 20.8%
- 27 azamat45/whisper-small-uz Whisper 20.8%
- 28 mrmuminov/whisper-tiny-uz Whisper 21%
- 29 blackhole33/whisper-small-uzb-commonvoice Whisper 22.2%
- 30 Inomjonov/mironshoh-whisper-medium-uz-10k Whisper 22.2%
- 31 mirodil/whisper-tiny-uzbek Whisper 22.6%
- 32 mirodil/whisper-small-uzbek Whisper 23.8%
- 33 Xojakbar/whisper-small-uz Whisper 39%
- 34 Saidakbar01/whisper-uz-finetuned Whisper 57%
Source: NavAI’s leaderboard
| # | Entry | Score | Setting | Submitted by | Date | Source |
|---|---|---|---|---|---|---|
| 1 | Kotib/uzbek_stt_v1 | 6.86% | Whisper / Training data undisclosed | Official leaderboard | ||
| 2 | nvidia/stt_uz_fastconformer_hybrid_large_pc | 7.01% | FastConformer / Single-pass decoding | Official leaderboard | ||
| 3 | uzinfocom-edu-ai/asr-uz-fastconformer-large | 7.06% | FastConformer / Single-pass decoding | Official leaderboard | ||
| 4 | navai-uz/whisper-medium-uzbek | 7.42% | Whisper | NavAI | Official leaderboard | |
| 5 | islomov/rubaistt_v2_medium | 7.64% | Whisper / FLEURS test data in training | Official leaderboard | ||
| 6 | navai-uz/whisper-small-uzbek | 7.92% | Whisper | NavAI | Official leaderboard | |
| 7 | OvozifyLabs/whisper-small-uz-v1 | 8.07% | Whisper | Official leaderboard | ||
| 8 | Abduqayum/whisper-uzbek-medium-callcenter | 8.28% | Whisper | Official leaderboard | ||
| 9 | navai-uz/whisper-base-uzbek | 10.3% | Whisper | NavAI | Official leaderboard | |
| 10 | aisha-org/Whisper-Uzbek | 10.6% | Whisper | Official leaderboard | ||
| 11 | Yorkinjon/whisper-medium-uzbek-ynv1 | 11.4% | Whisper | Official leaderboard | ||
| 12 | navai-uz/whisper-tiny-uzbek | 11.5% | Whisper | NavAI | Official leaderboard | |
| 13 | facebook/seamless-m4t-v2-large | 11.8% | Seamless / Single-pass decoding | Official leaderboard | ||
| 14 | mrmuminov/whisper-small-uz | 12% | Whisper | Official leaderboard | ||
| 15 | ai-sage/GigaAM-Multilingual (Large) | 12.9% | Conformer / Single-pass decoding | Official leaderboard | ||
| 16 | ai-sage/GigaAM-Multilingual (220M) | 13.4% | Conformer / Single-pass decoding | Official leaderboard | ||
| 17 | Yorkinjon/whisper-small-uzbek-ynv3 | 13.9% | Whisper | Official leaderboard | ||
| 18 | blackhole33/whisper-small-uz | 14.2% | Whisper | Official leaderboard | ||
| 19 | abduaziz/whisper-small-uzbek | 14.2% | Whisper | Official leaderboard | ||
| 20 | Yorkinjon/whisper-small-uzbek-ynv2 | 14.4% | Whisper | Official leaderboard | ||
| 21 | firdavsus/whisper-small-uz-ru | 14.8% | Whisper | Official leaderboard | ||
| 22 | mustafoyev202/whisper-uz | 14.8% | Whisper | Official leaderboard | ||
| 23 | Yorkinjon/whisper-small-uzbek-yn | 15.6% | Whisper | Official leaderboard | ||
| 24 | jmshd/whisper-uz | 16.2% | Whisper | Official leaderboard | ||
| 25 | tukhtashevshohruh/whisper-small-uz | 17.4% | Whisper | Official leaderboard | ||
| 26 | oyqiz/uzbek_stt | 20.8% | wav2vec2-CTC / Single-pass decoding | Official leaderboard | ||
| 27 | azamat45/whisper-small-uz | 20.8% | Whisper | Official leaderboard | ||
| 28 | mrmuminov/whisper-tiny-uz | 21% | Whisper | Official leaderboard | ||
| 29 | blackhole33/whisper-small-uzb-commonvoice | 22.2% | Whisper | Official leaderboard | ||
| 30 | Inomjonov/mironshoh-whisper-medium-uz-10k | 22.2% | Whisper | Official leaderboard | ||
| 31 | mirodil/whisper-tiny-uzbek | 22.6% | Whisper | Official leaderboard | ||
| 32 | mirodil/whisper-small-uzbek | 23.8% | Whisper | Official leaderboard | ||
| 33 | Xojakbar/whisper-small-uz | 39% | Whisper | Official leaderboard | ||
| 34 | Saidakbar01/whisper-uz-finetuned | 57% | 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 The tasks are private.
- 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 NavAI: 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.