Benchmark Almanac

LIMMITS 2024

TalTech 4 tops its leaderboard
Not in model cards
Version 2024

Speech models read text in a target voice across seven languages. Native listeners score naturalness and speaker similarity on the same prompts.

NVIDIA entered P-Flow in the zero-shot track. The organizers ranked seven systems there and specified 144 utterances per listening evaluation.

Released
September 2023
Kind
Audio understanding and voice
Level
Everyday
Scored
Graded by people
Answers
Open-ended
Tasks
All public
Measures
Capability
Language
Many languages
In and out
Text, Audio → Audio
Interaction
Single turn
Run by
Anyone
11
Entries on its own leaderboard, 0 models we track

TalTech tops the leaderboard with 4

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.

Bars start at 2.5.

  1. 1 TalTech Track 1: Few-shot, challenge data 4
  2. 2 NVIDIA Track 3: Zero-shot, external data allowed 4
  3. 3 SJTU_XLANCE_VC Track 2: Few-shot, external data allowed 4
  4. 4 THU_HCSI Track 1: Few-shot, challenge data 4
  5. 5 NLP_POSTECH Track 1: Few-shot, challenge data 4
  6. 6 LIMITLESS Track 2: Few-shot, external data allowed 4
  7. 7 MAIS_Speech_2024 Track 2: Few-shot, external data allowed 3
  8. 8 reply_2024 Track 3: Zero-shot, external data allowed 3
  9. 9 Shabdh Track 3: Zero-shot, external data allowed 3
  10. 10 MIDAS Track 1: Few-shot, challenge data 3
  11. 11 nwpu Track 1: Few-shot, challenge data 3

Source: Official LIMMITS 2024 leaderboard

#EntryScoreSettingSubmitted byDateLinkSource
1 TalTech 4 Track 1: Few-shot, challenge data TalTech docs.google.com Official leaderboard
2 NVIDIA 4 Track 3: Zero-shot, external data allowed NVIDIA docs.google.com Official leaderboard
3 SJTU_XLANCE_VC 4 Track 2: Few-shot, external data allowed SJTU_XLANCE_VC docs.google.com Official leaderboard
4 THU_HCSI 4 Track 1: Few-shot, challenge data THU_HCSI docs.google.com Official leaderboard
5 NLP_POSTECH 4 Track 1: Few-shot, challenge data NLP_POSTECH docs.google.com Official leaderboard
6 LIMITLESS 4 Track 2: Few-shot, external data allowed LIMITLESS docs.google.com Official leaderboard
7 MAIS_Speech_2024 3 Track 2: Few-shot, external data allowed MAIS_Speech_2024 docs.google.com Official leaderboard
8 reply_2024 3 Track 3: Zero-shot, external data allowed reply_2024 docs.google.com Official leaderboard
9 Shabdh 3 Track 3: Zero-shot, external data allowed Shabdh docs.google.com Official leaderboard
10 MIDAS 3 Track 1: Few-shot, challenge data MIDAS docs.google.com Official leaderboard
11 nwpu 3 Track 1: Few-shot, challenge data nwpu docs.google.com 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.

Versions

More in Audio and video

All 695 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 (11 entries).
  • Pass Models scored 11 entries on its leaderboard, none a model we track.
  • Note Newest score No scores yet.
  • Note Frontier models No models scored yet.
  • Note Headroom Scored as a rating, with no fixed ceiling.
Leaderboard import 3 pass
  • Pass Importer We read the board on sites.google.com: 11 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.

How we check