Benchmark Almanac

Voice Arena Hindi

ElevenLabs Eleven v4 1,108 tops its leaderboard Gemini 3.8 Flash TTS 1,084 leads the models we track
Gemini 3.8 Flash TTS 1,106 leads since September 2026
Only in system cards

Models generate speech in Hindi. Listeners compare anonymous clips, and preference votes determine independent language Elo ratings.

Each language has an independent leaderboard. Topic filters stay within its vote pool.

Made by
Voice Arena
Kind
Audio understanding and voice
Level
Everyday
Scored
Ranked by votes
Answers
Open-ended
Tasks
All public
Last in a system card
September 2026
Last model added
October 2026
Scores refreshed
November 12th, 2026
22
Entries on its own leaderboard, 9 models we track
2
System cards report it, from 1 lab

ElevenLabs Eleven v4 tops the leaderboard with 1,108

Best score per entry from the benchmark's own leaderboard, read on November 12th, 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. The best model we track is Gemini 3.8 Flash TTS, at 1,084.

Gemini 3.8 Flash TTS leads with 1,106, claims included

Best score per model from every source, including labs’ own numbers and claims Epoch AI collected from reports and papers. Labs pick their own settings, so a claim can beat an independent run of the same model.

22 entries. Scroll sideways for the rest. Bars start at 800.

  1. 1,108 ElevenLabs Eleven v4
  2. 1,103 Smallest AI Lightning 3.1 Pro (October)
  3. 1,086 Cartesia Sonic-3.6
  4. 1,084 Gemini 3.8 Flash TTS
  5. 1,070 Gemini 3.1 Flash TTS
  6. 1,065 Murf AI Falcon 2.1
  7. 1,057 Gemini 3.8 Flash-Lite TTS
  8. 1,037 Maya Research Maya-2-Native
  9. 1,037 Voxtral Mini TTS 3
  10. 1,035 ElevenLabs Eleven v3
  11. 1,027 Cartesia Sonic-3.5
  12. 1,016 Smallest AI Lightning 3.1 Pro
  13. 1,013 Murf AI Falcon 2
  14. 983 Mira-TTS
  15. 961 Chariot TTS-v0
  16. 953 Cartesia Sonic-3
  17. 948 Bulbul V3
  18. 933 MiniMax Speech 2.8 HD
  19. 930 TeenageWorks Teen-V1
  20. 883 Grok TTS
  21. 846 Dragon HD Omni
  22. 826 GPT-4o Mini TTS

Bars start at 800.

  1. 1 ElevenLabs Eleven v4 1,108
  2. 2 Smallest AI Lightning 3.1 Pro (October) 1,103
  3. 3 Cartesia Sonic-3.6 1,086
  4. 4 Gemini 3.8 Flash TTS 1,084
  5. 5 Gemini 3.1 Flash TTS 1,070
  6. 6 Murf AI Falcon 2.1 1,065
  7. 7 Gemini 3.8 Flash-Lite TTS 1,057
  8. 8 Maya Research Maya-2-Native 1,037
  9. 9 Voxtral Mini TTS 3 1,037
  10. 10 ElevenLabs Eleven v3 1,035
  11. 11 Cartesia Sonic-3.5 1,027
  12. 12 Smallest AI Lightning 3.1 Pro 1,016
  13. 13 Murf AI Falcon 2 1,013
  14. 14 Mira-TTS 983
  15. 15 Chariot TTS-v0 961
  16. 16 Cartesia Sonic-3 953
  17. 17 Bulbul V3 948
  18. 18 MiniMax Speech 2.8 HD 933
  19. 19 TeenageWorks Teen-V1 930
  20. 20 Grok TTS 883
  21. 21 Dragon HD Omni 846
  22. 22 GPT-4o Mini TTS 826

Source: Official Voice Arena Hindi leaderboard

Sources: Labs’ model cards and official Voice Arena Hindi leaderboard

#EntryScoreSettingDateSource
1 ElevenLabs Eleven v4 1,108 Official leaderboard
2 Smallest AI Lightning 3.1 Pro (October) 1,103 Official leaderboard
3 Cartesia Sonic-3.6 1,086 Official leaderboard
4 Gemini 3.8 Flash TTS 1,084 Sep 2026 Official leaderboard
5 Gemini 3.1 Flash TTS 1,070 Apr 2026 Official leaderboard
6 Murf AI Falcon 2.1 1,065 Official leaderboard
7 Gemini 3.8 Flash-Lite TTS 1,057 Sep 2026 Official leaderboard
8 Maya Research Maya-2-Native 1,037 Official leaderboard
9 Voxtral Mini TTS 3 1,037 Oct 2026 Official leaderboard
10 ElevenLabs Eleven v3 1,035 Official leaderboard
11 Cartesia Sonic-3.5 1,027 Official leaderboard
12 Smallest AI Lightning 3.1 Pro 1,016 Official leaderboard
13 Murf AI Falcon 2 1,013 Official leaderboard
14 Mira-TTS 983 Official leaderboard
15 Chariot TTS-v0 961 Official leaderboard
16 Cartesia Sonic-3 953 Official leaderboard
17 Bulbul V3 948 Feb 2026 Official leaderboard
18 MiniMax Speech 2.8 HD 933 Jan 2026 Official leaderboard
19 TeenageWorks Teen-V1 930 Official leaderboard
20 Grok TTS 883 Apr 2026 Official leaderboard
21 Dragon HD Omni 846 Dec 2025 Official leaderboard
22 GPT-4o Mini TTS 826 Mar 2025 Official leaderboard
#ModelScoreSettingDateSource
1 Gemini 3.8 Flash TTS 1,106 Sep 2026 Lab model card
2 Gemini 3.8 Flash-Lite TTS 1,076 Sep 2026 Lab model card
3 Gemini 3.1 Flash TTS 1,070 Apr 2026 Official leaderboard
4 Voxtral Mini TTS 3 1,037 Oct 2026 Official leaderboard
5 Bulbul V3 948 Feb 2026 Official leaderboard
6 MiniMax Speech 2.8 HD 933 Jan 2026 Official leaderboard
7 Grok TTS 883 Apr 2026 Official leaderboard
8 Dragon HD Omni 846 Dec 2025 Official leaderboard
9 GPT-4o Mini TTS 826 Mar 2025 Official leaderboard

Labs report scores for 2 models in their cards. Switch the scores above to rank them together.

6 models have held number one on the official leaderboard since 2023 6 models have held number one with claims included since 2023

Lab claims count from the day the card came out, independent runs from the model’s release.

OpenAI Microsoft MiniMax Sarvam AI Google DeepMind
Own leaderboard 6 leaders
Gemini 3.8 Flash TTS 1,084

Source: Official Voice Arena Hindi leaderboard

Scores over time

All the scores we have, one dot each, so a model tested at several settings appears more than once.

On the benchmark’s own leaderboard Reported by the lab in its model card
800 900 1,000 1,100 1,200 2025 2026 2027 800 900 1,000 1,100 1,200 2025 2026 2027

Sources: Official Voice Arena Hindi leaderboard and labs’ model cards

Labs report it only in system cards

We read the headline results of 522 launch posts from 52 labs, and none of them list this benchmark. Some system cards do, below.

2 system cards report it, from 1 lab

Besides launch posts, we read the system cards, model cards and technical reports labs publish with a release. They report far more, including safety and frontier-risk tests.

DateDocumentScoreAs the document writes itSection
Sep 23, 2026 Gemini 3.8 Flash TTS model card 1106 Voice Arena hindi, Elo, Gemini speech generation model page Capabilities
Sep 23, 2026 Gemini 3.8 Flash-Lite TTS model card 1076 Voice Arena hindi, Elo, Gemini speech generation model page Capabilities

2 lab claims next to the leaderboard

Each score a lab reported for its own model, beside the same model's best score on the leaderboard. Gaps mostly come from different harnesses, effort levels and task subsets. Every benchmark.

Model Lab says Board says Gap
Gemini 3.8 Flash TTS 1,106 Voice Arena hindi, Elo, Gemini speech generation model page 1,084 Leaderboard +22
Gemini 3.8 Flash-Lite TTS 1,076 Voice Arena hindi, Elo, Gemini speech generation model page 1,057 Leaderboard +19

More in Audio and video

All 233 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%