Benchmark Almanac

TTS Arena

Aurora 1,687 tops its leaderboard MiniMax Speech 2.8 HD 1,545 leads the models we track
Not in model cards

Users listen to two anonymous text-to-speech models read the same text and vote for the more natural one. Votes become Elo ratings.

The original space now points to TTS Arena V2, which has about 985 likes and was updated on October 31st, 2026. Release month of the first arena not confirmed on a maker page.

Released
February 2024
Kind
Audio understanding and voice
Level
Everyday
Scored
Ranked by votes
Answers
Open-ended
Tasks
All public
Last model added
January 2026
Scores refreshed
November 12th, 2026
45
Entries on its own leaderboard, 2 models we track

Aurora tops the leaderboard with 1,687

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 MiniMax Speech 2.8 HD, at 1,545.

45 entries. Scroll sideways for the rest. Bars start at 1,300.

  1. 1,687 Aurora
  2. 1,633 star-june-2026
  3. 1,602 Audio8-TTS-Realistic-0.6B
  4. 1,588 Gradium TTS 2026.10
  5. 1,567 Inworld TTS MAX
  6. 1,565 Inworld TTS 1.5 MAX
  7. 1,558 CastleFlow v1.0
  8. 1,554 Gradium TTS (pre-2026.08)
  9. 1,552 Papla P1
  10. 1,551 Hume Octave
  11. 1,545 MiniMax Speech 2.8 HD
  12. 1,544 Lightning v3.1 Pro
  13. 1,543 Inworld TTS
  14. 1,540 U2-TTS
  15. 1,539 OpenAudio S2
  16. 1,532 Deepdub eTTS 3.2
  17. 1,525 MiniMax Speech 2.6 Turbo
  18. 1,521 Eleven v3
  19. 1,515 SwanTale
  20. 1,507 Eleven Multilingual v2
  21. 1,505 Eleven Turbo v2.5
  22. 1,504 MiniMax Speech 2.6 HD
  23. 1,503 Eleven Flash v2.5
  24. 1,503 MiniMax Speech 02 HD
  25. 1,502 MiniMax Speech 2.8 Turbo
  26. 1,483 MiniMax Speech 02 Turbo
  27. 1,481 Parmesan
  28. 1,472 Typecast SSFM 3.0
  29. 1,470 Kokoro v1.0
  30. 1,461 Cartesia Sonic 2
  31. 1,461 Chatterbox
  32. 1,457 Magpie Research Preview
  33. 1,456 OpenAudio S1
  34. 1,454 Hithink Speech 2.6
  35. 1,436 Fluxions Vui
  36. 1,416 Silk Mulberry 1.5
  37. 1,400 Voice.ai Text to Speech V1
  38. 1,390 Tontaube V1
  39. 1,389 NeuTTS Max
  40. 1,383 Maya 1
  41. 1,372 Magpie Multilingual
  42. 1,344 Veena
  43. 1,332 Wordcab TTS
  44. 1,330 Vocu V3.0
  45. 1,329 Tontaube V0

Bars start at 1,300.

  1. 1 Aurora v2 1,687
  2. 2 star-june-2026 v2 1,633
  3. 3 Audio8-TTS-Realistic-0.6B v2 1,602
  4. 4 Gradium TTS 2026.10 v2 1,588
  5. 5 Inworld TTS MAX v2 1,567
  6. 6 Inworld TTS 1.5 MAX v2 1,565
  7. 7 CastleFlow v1.0 v2 1,558
  8. 8 Gradium TTS (pre-2026.08) v2 1,554
  9. 9 Papla P1 v2 1,552
  10. 10 Hume Octave v2 1,551
  11. 11 MiniMax Speech 2.8 HD v2 1,545
  12. 12 Lightning v3.1 Pro v2 1,544
  13. 13 Inworld TTS v2 1,543
  14. 14 U2-TTS v2 1,540
  15. 15 OpenAudio S2 v2 1,539
  16. 16 Deepdub eTTS 3.2 v2 1,532
  17. 17 MiniMax Speech 2.6 Turbo v2 1,525
  18. 18 Eleven v3 v2 1,521
  19. 19 SwanTale v2 1,515
  20. 20 Eleven Multilingual v2 v2 1,507
  21. 21 Eleven Turbo v2.5 v2 1,505
  22. 22 MiniMax Speech 2.6 HD v2 1,504
  23. 23 Eleven Flash v2.5 v2 1,503
  24. 24 MiniMax Speech 02 HD v2 1,503
  25. 25 MiniMax Speech 2.8 Turbo v2 1,502
  26. 26 MiniMax Speech 02 Turbo v2 1,483
  27. 27 Parmesan v2 1,481
  28. 28 Typecast SSFM 3.0 v2 1,472
  29. 29 Kokoro v1.0 v2 1,470
  30. 30 Cartesia Sonic 2 v2 1,461
  31. 31 Chatterbox v2 1,461
  32. 32 Magpie Research Preview v2 1,457
  33. 33 OpenAudio S1 v2 1,456
  34. 34 Hithink Speech 2.6 v2 1,454
  35. 35 Fluxions Vui v2 1,436
  36. 36 Silk Mulberry 1.5 v2 1,416
  37. 37 Voice.ai Text to Speech V1 v2 1,400
  38. 38 Tontaube V1 v2 1,390
  39. 39 NeuTTS Max v2 1,389
  40. 40 Maya 1 v2 1,383
  41. 41 Magpie Multilingual v2 1,372
  42. 42 Veena v2 1,344
  43. 43 Wordcab TTS v2 1,332
  44. 44 Vocu V3.0 v2 1,330
  45. 45 Tontaube V0 v2 1,329

Source: Official TTS Arena leaderboard

#EntryScoreSettingDateSource
1 Aurora 1,687 v2 Official leaderboard
2 star-june-2026 1,633 v2 Official leaderboard
3 Audio8-TTS-Realistic-0.6B 1,602 v2 Official leaderboard
4 Gradium TTS 2026.10 1,588 v2 Official leaderboard
5 Inworld TTS MAX 1,567 v2 Official leaderboard
6 Inworld TTS 1.5 MAX 1,565 v2 Official leaderboard
7 CastleFlow v1.0 1,558 v2 Official leaderboard
8 Gradium TTS (pre-2026.08) 1,554 v2 Official leaderboard
9 Papla P1 1,552 v2 Official leaderboard
10 Hume Octave 1,551 v2 Official leaderboard
11 MiniMax Speech 2.8 HD 1,545 v2 Jan 2026 Official leaderboard
12 Lightning v3.1 Pro 1,544 v2 Official leaderboard
13 Inworld TTS 1,543 v2 Official leaderboard
14 U2-TTS 1,540 v2 Official leaderboard
15 OpenAudio S2 1,539 v2 Official leaderboard
16 Deepdub eTTS 3.2 1,532 v2 Official leaderboard
17 MiniMax Speech 2.6 Turbo 1,525 v2 Official leaderboard
18 Eleven v3 1,521 v2 Official leaderboard
19 SwanTale 1,515 v2 Official leaderboard
20 Eleven Multilingual v2 1,507 v2 Official leaderboard
21 Eleven Turbo v2.5 1,505 v2 Official leaderboard
22 MiniMax Speech 2.6 HD 1,504 v2 Official leaderboard
23 Eleven Flash v2.5 1,503 v2 Official leaderboard
24 MiniMax Speech 02 HD 1,503 v2 Official leaderboard
25 MiniMax Speech 2.8 Turbo 1,502 v2 Jan 2026 Official leaderboard
26 MiniMax Speech 02 Turbo 1,483 v2 Official leaderboard
27 Parmesan 1,481 v2 Official leaderboard
28 Typecast SSFM 3.0 1,472 v2 Official leaderboard
29 Kokoro v1.0 1,470 v2 Official leaderboard
30 Cartesia Sonic 2 1,461 v2 Official leaderboard
31 Chatterbox 1,461 v2 Official leaderboard
32 Magpie Research Preview 1,457 v2 Official leaderboard
33 OpenAudio S1 1,456 v2 Official leaderboard
34 Hithink Speech 2.6 1,454 v2 Official leaderboard
35 Fluxions Vui 1,436 v2 Official leaderboard
36 Silk Mulberry 1.5 1,416 v2 Official leaderboard
37 Voice.ai Text to Speech V1 1,400 v2 Official leaderboard
38 Tontaube V1 1,390 v2 Official leaderboard
39 NeuTTS Max 1,389 v2 Official leaderboard
40 Maya 1 1,383 v2 Official leaderboard
41 Magpie Multilingual 1,372 v2 Official leaderboard
42 Veena 1,344 v2 Official leaderboard
43 Wordcab TTS 1,332 v2 Official leaderboard
44 Vocu V3.0 1,330 v2 Official leaderboard
45 Tontaube V0 1,329 v2 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
1,500 1,510 1,520 1,530 1,540 1,550 2026 1,500 1,510 1,520 1,530 1,540 1,550 2026

Source: Official TTS Arena 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%