Benchmark Almanac

Voice Arena Indian English

Gemini 3.8 Flash-Lite TTS 1,066
Not in model cards

Models generate speech in Indian English. 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 model added
September 2026
Scores refreshed
November 12th, 2026
7
Entries on its own leaderboard, 2 models we track

Gemini 3.8 Flash-Lite TTS leads with 1,066

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.

Bars start at 800.

  1. 1,066 Gemini 3.8 Flash-Lite TTS
  2. 1,062 Gemini 3.8 Flash TTS
  3. 1,059 ElevenLabs Eleven v4 Turbo
  4. 1,030 Inworld Realtime TTS 2
  5. 1,013 Cartesia Sonic-3.6
  6. 943 Gradium-TTS
  7. 827 Fish Audio Drama 3

Bars start at 800.

  1. 1 Gemini 3.8 Flash-Lite TTS 1,066
  2. 2 Gemini 3.8 Flash TTS 1,062
  3. 3 ElevenLabs Eleven v4 Turbo 1,059
  4. 4 Inworld Realtime TTS 2 1,030
  5. 5 Cartesia Sonic-3.6 1,013
  6. 6 Gradium-TTS 943
  7. 7 Fish Audio Drama 3 827

Source: Official Voice Arena Indian English leaderboard

#EntryScoreSettingDateSource
1 Gemini 3.8 Flash-Lite TTS 1,066 Sep 2026 Official leaderboard
2 Gemini 3.8 Flash TTS 1,062 Sep 2026 Official leaderboard
3 ElevenLabs Eleven v4 Turbo 1,059 Official leaderboard
4 Inworld Realtime TTS 2 1,030 Official leaderboard
5 Cartesia Sonic-3.6 1,013 Official leaderboard
6 Gradium-TTS 943 Official leaderboard
7 Fish Audio Drama 3 827 Official 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
1,062 1,063 1,064 1,065 1,066 1,062 1,063 1,064 1,065 1,066

Source: Official Voice Arena Indian English leaderboard

No lab model card reports it yet

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

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%