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 | 1-2 |  Alibaba | Qwen-Audio-3.0-TTS-Plus | 1,237 | -17/17 | 1,427 | Jul 2026 | $27.6 /1M chars | |
| 2 | 1-2 |  SpeechifyAI | Simba 3.2 | 1,232 | -16/16 | 1,405 | Jul 2026 | $10.0 /1M chars | |
| 3 | 3-6 |  Google | Gemini 3.1 Flash TTS | 1,211 | -14/14 | 2,775 | Apr 2026 | $18.3 /1M chars | |
| 4 | 3-6 |  Cartesia | Sonic 3.5 | 1,211 | -14/14 | 2,238 | May 2026 | $49.0 /1M chars | |
| 5 | 3-7 |  Inworld | Realtime TTS 1.5 Max | 1,200 | -14/14 | 2,007 | Jan 2026 | $26.0 /1M chars | |
| 6 | 3-7 | Inworld | Realtime TTS-2 - Research Preview | 1,200 | -14/14 | 1,998 | May 2026 | $20.8 /1M chars | |
| 7 | 5-11 |  VUI Labs | Luna TTS | 1,186 | -16/16 | 1,282 | Jun 2026 | N/A | |
| 8 | 7-11 |  MiniMax | Speech 2.8 HD | 1,180 | -12/12 | 3,876 | Feb 2026 | $100.0 /1M chars | |
| 9 | 7-11 |  StepFun | StepAudio 2.5 TTS | 1,175 | -18/18 | 1,031 | Apr 2026 | $85.0 /1M chars | |
| 10 | 7-11 |  async | Async Flash v1.5 | 1,173 | -15/15 | 1,421 | May 2026 | $10.1 /1M chars | |
| 11 | 8-11 |  ElevenLabs | Eleven v3 | 1,173 | -12/12 | 3,742 | Feb 2026 | $100.0 /1M chars | |
| 12 | 12-16 | MiniMax | Speech 2.8 Turbo | 1,151 | -12/12 | 3,790 | Feb 2026 | $60.0 /1M chars | |
| 13 | 12-18 |  Smallest.ai | Lightning V3.1 Pro TTS (Jun 2026) | 1,149 | -15/15 | 1,332 | Jun 2026 | $19.5 /1M chars | |
| 14 | 12-18 | StepFun | Step TTS 2 (Mar 2026) | 1,146 | -14/14 | 1,644 | Feb 2026 | $40.0 /1M chars | |
| 15 | 12-19 | async | Async Pro v1.0 | 1,144 | -15/15 | 1,378 | May 2026 | $20.2 /1M chars | |
| 16 | 12-19 |  Fish Audio | Fish Audio S2.1 Pro | 1,139 | -15/15 | 1,367 | Jun 2026 | $15.0 /1M chars | |
| 17 | 13-19 | MiniMax | Speech 2.6 HD | 1,138 | -12/12 | 3,539 | Oct 2025 | $100.0 /1M chars | |
| 18 | 13-19 | Inworld | Realtime TTS 1.5 Mini | 1,137 | -13/13 | 2,492 | Jan 2026 | $10.4 /1M chars | |
| 19 | 16-23 | MiniMax | Speech 2.6 Turbo | 1,129 | -12/12 | 3,897 | Oct 2025 | $60.0 /1M chars | |
| 20 | 18-24 |  Microsoft | Azure HD 2.5 | 1,124 | -14/14 | 1,885 | Mar 2026 | $22.0 /1M chars | |
| 21 | 19-25 | SpeechifyAI | Simba 3.0 | 1,121 | -14/14 | 1,810 | Feb 2026 | $12.5 /1M chars | |
| 22 | 15-28 |  SpaceXAI | SpaceXAI TTS | 1,121 | -24/24 | 296 | Mar 2026 | $15.0 /1M chars | |
| 23 | 19-26 | Fish Audio | Fish Audio S2 Pro  Open Weights | 1,117 | -14/14 | 1,505 | Mar 2026 | $15.0 /1M chars | |
| 24 | 20-28 | MiniMax | Speech-02-HD | 1,112 | -12/12 | 4,016 | Mar 2025 | $100.0 /1M chars | |
| 25 | 20-28 | StepFun | Step Audio EditX (Mar 2026) Open Weights | 1,109 | -15/15 | 1,419 | Feb 2026 | N/A | |
| 26 | 24-28 |  OpenAI | TTS-1 HD | 1,104 | -12/12 | 3,158 | Nov 2023 | $30.0 /1M chars | |
| 27 | 24-28 | ElevenLabs | Multilingual v2 | 1,103 | -11/11 | 7,375 | Aug 2023 | $100.0 /1M chars | |
| 28 | 24-28 | ElevenLabs | Turbo v2.5 | 1,103 | -11/11 | 6,924 | Jul 2024 | $50.0 /1M chars | |
| 29 | 26-34 |  Gradium | Gradium TTS | 1,090 | -15/15 | 1,178 | Jun 2026 | N/A | |
| 30 | 29-34 | ElevenLabs | Flash v2.5 | 1,088 | -11/11 | 5,605 | Dec 2024 | $50.0 /1M chars | |
| 31 | 29-35 | MiniMax | Speech-02-Turbo | 1,085 | -12/12 | 3,535 | Mar 2025 | $60.0 /1M chars | |
| 32 | 29-35 | OpenAI | TTS-1 | 1,084 | -11/11 | 6,087 | Nov 2023 | $15.0 /1M chars | |
| 33 | 29-38 |  Resemble AI | Chatterbox HD | 1,083 | -14/14 | 1,963 | — | May 2025 | $40.0 /1M chars |
| 34 | 29-42 |  Mistral | Voxtral TTS Open Weights | 1,079 | -14/14 | 1,519 | Mar 2026 | $16.0 /1M chars | |
| 35 | 31-42 | Google | Gemini 2.5 Flash Lite TTS | 1,074 | -13/13 | 2,616 | Dec 2025 | $9.2 /1M chars | |
| 36 | 31-45 | OpenAI | GPT-Realtime-2 | 1,072 | -15/15 | 1,189 | May 2026 | $191.6 /1M chars | |
| 37 | 33-44 | Cartesia | Sonic 3 | 1,071 | -12/12 | 3,157 | Oct 2025 | $49.0 /1M chars | |
| 38 | 34-44 | Fish Audio | OpenAudio S1 | 1,070 | -12/12 | 4,221 | Jun 2025 | $15.0 /1M chars | |
| 39 | 34-45 | Google | Studio | 1,068 | -12/12 | 4,266 | Feb 2023 | $160.0 /1M chars | |
| 40 | 34-45 |  Amazon | Polly Generative | 1,067 | -12/12 | 3,532 | May 2024 | $30.0 /1M chars | |
| 41 | 34-47 |  Xiaomi | MiMo-V2.5-TTS | 1,067 | -15/15 | 1,361 | - | N/A | |
| 42 | 34-47 | SpeechifyAI | SIMBA 1.6 | 1,065 | -14/14 | 1,796 | Oct 2024 | $10.0 /1M chars | |
| 43 | 36-48 | MiniMax | T2A-01-HD | 1,061 | -12/12 | 4,737 | Jan 2025 | $50.0 /1M chars | |
| 44 | 36-51 |  NVIDIA | Magpie-Multilingual 357M (Feb 2026) Open Weights | 1,060 | -14/14 | 1,444 | Feb 2026 | N/A | |
| 45 | 39-51 |  Kokoro | Kokoro 82M v1.0 Open Weights | 1,058 | -11/11 | 5,208 | Jan 2025 | $0.7 /1M chars | |
| 46 | 40-53 |  Hume AI | Octave 2 | 1,055 | -13/13 | 2,821 | Oct 2025 | $87.5 /1M chars | |
| 47 | 43-54 | Google | Chirp 3: HD | 1,052 | -12/12 | 3,071 | Mar 2025 | $30.0 /1M chars | |
| 48 | 42-56 |  Hithink | Speech 2.6 | 1,050 | -16/16 | 1,137 | Jul 2026 | N/A | |
| 49 | 44-54 | async | Async Flash v1.0 | 1,049 | -12/12 | 3,646 | Jul 2025 | $10.1 /1M chars | |
| 50 | 44-55 |  Maya Research | Maya1 Open Weights | 1,048 | -13/13 | 3,025 | Jun 2025 | N/A | |
| 51 | 43-57 | Amazon | Polly Long-Form | 1,047 | -14/14 | 1,613 | Nov 2023 | $100.0 /1M chars | |
| 52 | 43-60 | Fish Audio | OpenAudio S1 Mini | 1,045 | -20/20 | 1,777 | Jun 2025 | $15.0 /1M chars | |
| 53 | 46-58 | Google | Journey | 1,043 | -14/14 | 1,846 | Dec 2023 | $160.0 /1M chars | |
| 54 | 47-58 | Cartesia | Sonic English (Oct 2024) | 1,041 | -12/12 | 3,173 | Oct 2024 | $49.0 /1M chars | |
| 55 | 49-62 |  Rime | Coda | 1,036 | -14/14 | 1,482 | May 2026 | $50.0 /1M chars | |
| 56 | 50-63 | Google | Gemini 2.5 Pro (Dec 2025) | 1,034 | -14/14 | 2,037 | Dec 2025 | $18.3 /1M chars | |
| 57 | 52-63 | Google | Gemini 2.5 Flash TTS (Dec 2025) | 1,034 | -13/13 | 2,413 | Dec 2025 | $9.2 /1M chars | |
| 58 | 53-63 | SpeechifyAI | SIMBA 1.0 | 1,033 | -12/12 | 4,716 | Jun 2024 | $10.0 /1M chars | |
| 59 | 52-65 | Microsoft | Azure Neural | 1,028 | -19/19 | 1,222 | Sept 2018 | $15.0 /1M chars | |
| 60 | 55-64 | Microsoft | MAI-Voice-1 | 1,026 | -14/14 | 1,667 | Apr 2026 | N/A | |
| 61 | 55-65 | Smallest.ai | Lightning v3.1 | 1,024 | -14/14 | 1,802 | Mar 2026 | $25.0 /1M chars | |
| 62 | 56-65 | Hume AI | Octave TTS | 1,024 | -12/12 | 4,655 | Feb 2025 | $87.5 /1M chars | |
| 63 | 58-65 | MiniMax | T2A-01-Turbo | 1,021 | -12/12 | 4,881 | Jan 2025 | $30.0 /1M chars | |
| 64 | 61-67 | Resemble AI | Chatterbox  Open Weights | 1,013 | -12/12 | 4,534 | May 2025 | $25.0 /1M chars | |
| 65 | 60-68 | Xiaomi | MiMo-V2-TTS | 1,012 | -15/15 | 1,331 | Mar 2026 | N/A | |
| 66 | 64-68 | NVIDIA | Magpie-Multilingual 357M Open Weights | 1,006 | -13/13 | 3,379 | Jan 2026 | N/A | |
| 67 | 64-68 | Rime | Arcana v3 | 1,003 | -14/14 | 1,735 | Feb 2026 | $50.0 /1M chars | |
| 68 | 68 |  Zyphra | Zonos-v0.1 Open Weights | 1,000 | 0/0 | 4,766 | Feb 2025 | $20.0 /1M chars | |
| 69 | 69-71 |  Murf AI | Murf Speech Gen 2 | 970 | -12/12 | 5,046 | Jul 2024 | $100.0 /1M chars | |
| 70 | 69-73 |  LMNT | LMNT | 968 | -13/13 | 3,838 | Dec 2023 | $49.0 /1M chars | |
| 71 | 69-73 | Microsoft | VibeVoice 1.5B | 967 | -14/14 | 1,892 | Aug 2025 | N/A | |
| 72 | 69-73 |  OpenVoice | OpenVoice v2 Open Weights | 957 | -14/14 | 2,866 | — | Apr 2024 | $8.3 /1M chars |
| 73 | 70-73 | Microsoft | VibeVoice 7B Open Weights | 956 | -14/14 | 1,855 | Aug 2025 | N/A | |
| 74 | 74-76 | NVIDIA | Magpie Multilingual | 939 | -15/15 | 1,741 | Apr 2025 | N/A | |
| 75 | 74-77 |  Neuphonic | Neuphonic TTS | 934 | -14/14 | 2,077 | Oct 2024 | $20.8 /1M chars | |
| 76 | 74-77 | Alibaba | Qwen3 TTS Flash | 929 | -15/15 | 1,840 | Sept 2025 | $10.0 /1M chars | |
| 77 | 75-78 | Alibaba | Qwen3 TTS | 921 | -14/14 | 2,446 | - | N/A | |
| 78 | 77-79 |  Coqui | XTTS v2 Open Weights | 914 | -15/15 | 2,348 | — | Nov 2023 | $40.4 /1M chars |
| 79 | 78-79 | Google | WaveNet | 903 | -12/12 | 6,404 | Sept 2016 | $16.0 /1M chars | |
| 80 | 79-84 |  StyleTTS | StyleTTS 2 Open Weights | 888 | -16/16 | 2,283 | — | Jun 2023 | $2.8 /1M chars |
| 81 | 80-84 | Rime | Mist V2 | 886 | -15/15 | 2,642 | Mar 2025 | $31.1 /1M chars | |
| 82 | 80-84 | Google | Neural2 | 883 | -12/12 | 6,640 | Jun 2022 | $16.0 /1M chars | |
| 83 | 80-84 | Amazon | Polly Neural | 879 | -15/15 | 2,622 | Jul 2019 | $16.0 /1M chars | |
| 84 | 81-84 | Google | Standard | 876 | -12/12 | 6,753 | Aug 2018 | $4.0 /1M chars | |
| 85 | 85-86 |  Noiz | Noiz TTS | 852 | -17/17 | 1,324 | Nov 2025 | N/A | |
| 86 | 85-86 | Murf AI | Falcon (Beta) | 849 | -15/15 | 2,699 | Nov 2025 | N/A | |
| 87 | 86-87 |  MetaVoice | MetaVoice v1 Open Weights | 833 | -18/18 | 1,255 | — | Feb 2024 | N/A |
| 88 | 88 | Amazon | Polly Standard | 811 | -16/16 | 3,452 | 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
Qwen-Audio-3.0-TTS-Plus currently leads the Text to Speech Arena with an Elo score of 1237.
The top Text to Speech models by Elo rating are: 1. Qwen-Audio-3.0-TTS-Plus (Elo 1237), 2. Simba 3.2 (Elo 1232), 3. Gemini 3.1 Flash TTS (Elo 1211), 4. Sonic 3.5 (Elo 1211), 5. Realtime TTS 1.5 Max (Elo 1200). Rankings are based on blind user votes in the Speech Arena.
Models are ranked using an Elo rating system derived from user votes in blind comparisons in the Speech Arena. Users listen to pairs of speech samples generated from the same text and choose which sounds more natural. Higher Elo scores indicate a model produces speech preferred more often by listeners. Vote 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.
Fish Audio S2 Pro is the highest-ranked open weights model on the Text to Speech Leaderboard with an Elo score of 1117. There are 14 open weights models out of 88 total.
The top 5 open weights Text to Speech models are: 1. Fish Audio S2 Pro (Elo 1117), 2. Step Audio EditX (Mar 2026) (Elo 1109), 3. Voxtral TTS (Elo 1079), 4. Magpie-Multilingual 357M (Feb 2026) (Elo 1060), 5. Kokoro 82M v1.0 (Elo 1058).
You can filter by the following categories: Knowledge Sharing, Assistants, Entertainment, and Customer Service, and the following accents: US and UK.