Speech Arena LeaderboardArtificial Analysis
Compare TTS models using each provider's own native voices.
Range | Creator | Model | Elo | 95% CI | Samples | Arena Voices | Released | API Pricing 1 | |
|---|---|---|---|---|---|---|---|---|---|
| 1 | 1-3 | Simba 3.2 | 1,232 | -15/15 | 1,911 | Jul 2026 | $10.0 /1M chars | ||
| 2 | 1-3 | Qwen-Audio-3.0-TTS-Plus | 1,229 | -15/15 | 1,890 | Jul 2026 | $27.6 /1M chars | ||
| 3 | 1-5 | Luna TTS | 1,218 | -15/15 | 1,931 | Jun 2026 | $80.0 /1M chars | ||
| 4 | 3-6 | Gemini 3.1 Flash TTS | 1,209 | -13/13 | 3,236 | Apr 2026 | $18.3 /1M chars | ||
| 5 | 3-8 | StepAudio 2.5 TTS | 1,205 | -15/15 | 1,590 | Jun 2026 | $85.0 /1M chars | ||
| 6 | 4-8 | Sonic 3.5 | 1,202 | -13/13 | 2,703 | May 2026 | $49.0 /1M chars | ||
| 7 | 5-9 | Realtime TTS 1.5 Max | 1,192 | -13/13 | 2,349 | Jan 2026 | $26.0 /1M chars | ||
| 8 | 5-10 | Lightning V3.1 Pro (Jul 2026) | 1,191 | -16/16 | 1,453 | Jul 2026 | $19.5 /1M chars | ||
| 9 | 7-10 | Realtime TTS-2 - Research Preview | 1,188 | -13/13 | 2,412 | May 2026 | $20.8 /1M chars | ||
| 10 | 9-11 | Speech 2.8 HD | 1,177 | -12/12 | 4,245 | Feb 2026 | $100.0 /1M chars | ||
| 11 | 10-11 | Eleven v3 | 1,172 | -12/12 | 4,114 | Feb 2026 | $100.0 /1M chars | ||
| 12 | 12-14 | Speech 2.8 Turbo | 1,153 | -12/12 | 4,131 | Feb 2026 | $60.0 /1M chars | ||
| 13 | 12-18 | Gradium TTS Beta | 1,150 | -16/16 | 1,126 | Jul 2026 | $47.2 /1M chars | ||
| 14 | 12-17 | Async Flash v1.5 | 1,149 | -14/14 | 1,820 | May 2026 | $10.1 /1M chars | ||
| 15 | 12-20 | Step TTS 2 (Mar 2026) | 1,141 | -13/13 | 1,953 | Feb 2026 | $40.0 /1M chars | ||
| 16 | 12-21 | Fish Audio S2.1 Pro | 1,139 | -14/14 | 1,794 | Jun 2026 | $15.0 /1M chars | ||
| 17 | 12-21 | Lightning V3.1 Pro TTS (Jun 2026) | 1,139 | -14/14 | 1,760 | Jun 2026 | $19.5 /1M chars | ||
| 18 | 15-22 | Speech 2.6 HD | 1,134 | -12/12 | 3,792 | Oct 2025 | $100.0 /1M chars | ||
| 19 | 15-24 | Realtime TTS 1.5 Mini | 1,130 | -12/12 | 2,761 | Jan 2026 | $10.4 /1M chars | ||
| 20 | 15-24 | Async Pro v1.0 | 1,129 | -14/14 | 1,733 | May 2026 | $20.2 /1M chars | ||
| 21 | 18-24 | Azure HD 2.5 | 1,126 | -13/13 | 2,225 | Mar 2026 | $22.0 /1M chars | ||
| 22 | 18-25 | Speech 2.6 Turbo | 1,122 | -12/12 | 4,164 | Oct 2025 | $60.0 /1M chars | ||
| 23 | 18-25 | Fish Audio S2 Pro | 1,122 | -13/13 | 1,853 | Mar 2026 | $15.0 /1M chars | ||
| 24 | 19-27 | Simba 3.0 | 1,120 | -13/13 | 2,211 | Feb 2026 | $12.5 /1M chars | ||
| 25 | 22-30 | Speech-02-HD | 1,112 | -12/12 | 4,188 | Mar 2025 | $100.0 /1M chars | ||
| 26 | 24-30 | Step Audio EditX (Mar 2026) | 1,108 | -14/14 | 1,714 | Feb 2026 | N/A | ||
| 27 | 22-30 | SpaceXAI TTS | 1,107 | -18/18 | 721 | Mar 2026 | $15.0 /1M chars | ||
| 28 | 25-30 | TTS-1 HD | 1,101 | -12/12 | 3,362 | Nov 2023 | $30.0 /1M chars | ||
| 29 | 26-30 | Turbo v2.5 | 1,100 | -11/11 | 7,125 | Jul 2024 | $50.0 /1M chars | ||
| 30 | 26-30 | Multilingual v2 | 1,100 | -11/11 | 7,559 | Aug 2023 | $100.0 /1M chars | ||
| 31 | 31-37 | Chatterbox HD | 1,087 | -13/13 | 2,144 | — | May 2025 | $40.0 /1M chars | |
| 32 | 31-37 | TTS-1 | 1,086 | -11/11 | 6,282 | Nov 2023 | $15.0 /1M chars | ||
| 33 | 31-38 | Gradium TTS | 1,086 | -14/14 | 1,579 | Jun 2026 | $47.2 /1M chars | ||
| 34 | 31-41 | Flash v2.5 | 1,081 | -11/11 | 5,776 | Dec 2024 | $50.0 /1M chars | ||
| 35 | 31-42 | Speech-02-Turbo | 1,081 | -12/12 | 3,711 | Mar 2025 | $60.0 /1M chars | ||
| 36 | 31-42 | Gemini 2.5 Flash Lite TTS | 1,079 | -12/12 | 2,955 | Dec 2025 | $9.2 /1M chars | ||
| 37 | 31-44 | Studio | 1,077 | -12/12 | 4,464 | Feb 2023 | $160.0 /1M chars | ||
| 38 | 34-49 | Voxtral TTS | 1,073 | -13/13 | 1,882 | Mar 2026 | $16.0 /1M chars | ||
| 39 | 32-50 | Maya 2 Flash | 1,072 | -15/15 | 1,520 | Jul 2026 | N/A | ||
| 40 | 34-48 | OpenAudio S1 | 1,071 | -11/11 | 4,408 | Jun 2025 | $15.0 /1M chars | ||
| 41 | 34-50 | Speech 2.6 | 1,071 | -14/14 | 1,550 | Jul 2026 | N/A | ||
| 42 | 34-50 | Sonic 3 | 1,070 | -12/12 | 3,449 | Oct 2025 | $49.0 /1M chars | ||
| 43 | 37-51 | Polly Generative | 1,066 | -12/12 | 3,737 | May 2024 | $30.0 /1M chars | ||
| 44 | 37-52 | Magpie-Multilingual 357M (Feb 2026) | 1,065 | -13/13 | 1,787 | Feb 2026 | N/A | ||
| 45 | 37-53 | Journey | 1,064 | -13/13 | 2,049 | Dec 2023 | $160.0 /1M chars | ||
| 46 | 38-53 | SIMBA 1.6 | 1,063 | -13/13 | 2,103 | Oct 2024 | $10.0 /1M chars | ||
| 47 | 38-53 | T2A-01-HD | 1,062 | -11/11 | 4,915 | Jan 2025 | $50.0 /1M chars | ||
| 48 | 38-54 | GPT-Realtime-2 | 1,061 | -14/14 | 1,619 | May 2026 | $191.6 /1M chars | ||
| 49 | 38-54 | MiMo-V2.5-TTS | 1,060 | -13/13 | 1,741 | - | N/A | ||
| 50 | 43-54 | Kokoro 82M v1.0 | 1,058 | -11/11 | 5,385 | Jan 2025 | $0.7 /1M chars | ||
| 51 | 43-58 | Octave 2 | 1,055 | -12/12 | 2,981 | Oct 2025 | $87.5 /1M chars | ||
| 52 | 43-59 | Coda | 1,053 | -13/13 | 1,915 | May 2026 | $50.0 /1M chars | ||
| 53 | 46-60 | Chirp 3: HD | 1,051 | -12/12 | 3,342 | Mar 2025 | $30.0 /1M chars | ||
| 54 | 49-61 | Async Flash v1.0 | 1,048 | -12/12 | 3,954 | Jul 2025 | $10.1 /1M chars | ||
| 55 | 43-66 | OpenAudio S1 Mini | 1,046 | -20/20 | 1,777 | Jun 2025 | $15.0 /1M chars | ||
| 56 | 50-63 | Gemini 2.5 Flash TTS (Dec 2025) | 1,046 | -13/13 | 2,708 | Dec 2025 | $9.2 /1M chars | ||
| 57 | 49-64 | Maya 2 Global | 1,045 | -15/15 | 1,455 | Jul 2026 | N/A | ||
| 58 | 51-63 | Maya1 | 1,043 | -12/12 | 3,255 | Jun 2025 | N/A | ||
| 59 | 51-66 | Polly Long-Form | 1,041 | -14/14 | 1,795 | Nov 2023 | $100.0 /1M chars | ||
| 60 | 52-66 | Higgs Audio V3 TTS | 1,040 | -14/14 | 1,735 | Jun 2026 | N/A | ||
| 61 | 54-67 | Sonic English (Oct 2024) | 1,037 | -12/12 | 3,318 | Oct 2024 | $49.0 /1M chars | ||
| 62 | 57-68 | SIMBA 1.0 | 1,034 | -11/11 | 4,886 | Jun 2024 | $10.0 /1M chars | ||
| 63 | 55-68 | Gemini 2.5 Pro (Dec 2025) | 1,033 | -13/13 | 2,348 | Dec 2025 | $18.3 /1M chars | ||
| 64 | 59-68 | MAI-Voice-1 | 1,030 | -13/13 | 2,027 | Apr 2026 | N/A | ||
| 65 | 61-68 | T2A-01-Turbo | 1,028 | -11/11 | 5,037 | Jan 2025 | $30.0 /1M chars | ||
| 66 | 56-70 | Azure Neural | 1,027 | -19/19 | 1,226 | Sept 2018 | $15.0 /1M chars | ||
| 67 | 61-70 | Octave TTS | 1,025 | -12/12 | 4,792 | Feb 2025 | $87.5 /1M chars | ||
| 68 | 62-70 | Lightning v3.1 | 1,024 | -13/13 | 2,088 | Mar 2026 | $25.0 /1M chars | ||
| 69 | 65-72 | Chatterbox | 1,017 | -12/12 | 4,693 | May 2025 | $25.0 /1M chars | ||
| 70 | 65-72 | MiMo-V2-TTS | 1,014 | -14/14 | 1,621 | Mar 2026 | N/A | ||
| 71 | 69-73 | Magpie-Multilingual 357M | 1,005 | -12/12 | 3,644 | Jan 2026 | N/A | ||
| 72 | 69-73 | Arcana v3 | 1,005 | -13/13 | 2,086 | Feb 2026 | $50.0 /1M chars | ||
| 73 | 73 | Zonos-v0.1 | 1,000 | 0/0 | 4,908 | Feb 2025 | $20.0 /1M chars | ||
| 74 | 74-76 | Murf Speech Gen 2 | 976 | -12/12 | 5,193 | Jul 2024 | $100.0 /1M chars | ||
| 75 | 74-77 | VibeVoice 1.5B | 972 | -14/14 | 2,014 | Aug 2025 | N/A | ||
| 76 | 74-77 | LMNT | 970 | -13/13 | 3,957 | Dec 2023 | $49.0 /1M chars | ||
| 77 | 74-78 | VibeVoice 7B | 964 | -14/14 | 1,979 | Aug 2025 | N/A | ||
| 78 | 77-79 | OpenVoice v2 | 952 | -13/13 | 2,971 | — | Apr 2024 | $8.3 /1M chars | |
| 79 | 78-81 | Magpie Multilingual | 940 | -14/14 | 1,850 | Apr 2025 | N/A | ||
| 80 | 78-81 | Qwen3 TTS Flash | 939 | -14/14 | 1,970 | Sept 2025 | $10.0 /1M chars | ||
| 81 | 79-81 | Neuphonic TTS | 934 | -14/14 | 2,200 | Oct 2024 | $20.8 /1M chars | ||
| 82 | 82-84 | Qwen3 TTS | 919 | -13/13 | 2,679 | - | N/A | ||
| 83 | 82-84 | XTTS v2 | 917 | -15/15 | 2,449 | — | Nov 2023 | $40.4 /1M chars | |
| 84 | 83-84 | WaveNet | 907 | -12/12 | 6,496 | Sept 2016 | $16.0 /1M chars | ||
| 85 | 85-88 | Mist V2 | 891 | -15/15 | 2,736 | Mar 2025 | $31.1 /1M chars | ||
| 86 | 85-89 | StyleTTS 2 | 888 | -15/15 | 2,366 | — | Jun 2023 | $2.8 /1M chars | |
| 87 | 85-89 | Neural2 | 885 | -12/12 | 6,747 | Jun 2022 | $16.0 /1M chars | ||
| 88 | 85-89 | Polly Neural | 879 | -14/14 | 2,696 | Jul 2019 | $16.0 /1M chars | ||
| 89 | 87-89 | Standard | 875 | -12/12 | 6,841 | Aug 2018 | $4.0 /1M chars | ||
| 90 | 89-90 | Noiz TTS | 862 | -16/16 | 1,646 | Nov 2025 | N/A | ||
| 91 | 91 | MetaVoice v1 | 840 | -18/18 | 1,325 | — | Feb 2024 | N/A | |
| 92 | 92 | Polly Standard | 813 | -16/16 | 3,521 | Nov 2016 | $4.0 /1M chars |
1 API Pricing reflects the cost to generate 1M characters on the model creator's API with the model's default settings
Frequently Asked Questions
The top Text to Speech models by Elo rating are: 1. Simba 3.2 (Elo 1232), 2. Qwen-Audio-3.0-TTS-Plus (Elo 1229), 3. Luna TTS (Elo 1218), 4. Gemini 3.1 Flash TTS (Elo 1209), 5. StepAudio 2.5 TTS (Elo 1205). Rankings are based on blind user votes in the Speech Arena.
Kokoro 82M v1.0 is the most affordable at $0.65 per 1M characters with an Elo score of 1058. Other affordable options include StyleTTS 2 at $2.82 per 1M characters.
The top 5 open weights Text to Speech models are: 1. Fish Audio S2 Pro (Elo 1122), 2. Step Audio EditX (Mar 2026) (Elo 1108), 3. Voxtral TTS (Elo 1073), 4. Magpie-Multilingual 357M (Feb 2026) (Elo 1065), 5. Kokoro 82M v1.0 (Elo 1058).