Best AI Voice & Speech Models
See how the top voice models compare, when each launched, and which labs are ahead.
See how the top voice models compare, when each launched, and which labs are ahead.
ELO arena ranking for speech synthesis quality based on blind pairwise human preference comparisons. Data from Artificial Analysis
| # | ||
|---|---|---|
| 1 | SpeechifyAI | 1,229±16 |
| 2 | Alibaba | 1,227±16 |
| 3 | Google | 1,212±13 |
| 4 | Inworld | 1,212±19 |
| 5 | VUI Labs | 1,205±15 |
| 6 | Alibaba | 1,205±15 |
| 7 | Cartesia | 1,202±14 |
| 8 | StepFun | 1,199±16 |
| 9 | Inworld | 1,196±14 |
| 10 | Alibaba | 1,193±15 |
| 11 | Smallest.ai | 1,190±17 |
| 12 | Inworld | 1,189±14 |
| 13 | StepFun | 1,175±18 |
| 14 | ElevenLabs | 1,173±12 |
| 15 | MiniMax | 1,172±12 |
| 16 | Inworld | 1,161±13 |
| 17 | Inworld | 1,159±16 |
| 18 | async | 1,155±14 |
| 19 | MiniMax | 1,146±12 |
| 20 | Fish Audio | 1,139±14 |
| 21 | StepFun | 1,139±14 |
| 22 | Smallest.ai | 1,137±14 |
| 23 | Inworld | 1,132±13 |
| 24 | MiniMax | 1,132±12 |
| 25 | async | 1,128±14 |
| 26 | Fish Audio | 1,123±14 |
| 27 | Microsoft Azure | 1,122±13 |
| 28 | MiniMax | 1,122±12 |
| 29 | Speechify | 1,121±13 |
| 30 | Inworld | 1,121±12 |
| 31 | MiniMax | 1,109±12 |
| 32 | StepFun | 1,108±14 |
| 33 | StepFun | 1,105±19 |
| 34 | OpenAI | 1,102±12 |
| 35 | SpaceXAI | 1,101±19 |
| 36 | ElevenLabs | 1,098±11 |
| 37 | ElevenLabs | 1,097±11 |
| 38 | ElevenLabs | 1,095±11 |
| 39 | Gradium | 1,094±16 |
| 40 | Smallest.ai | 1,090±14 |
| 41 | Resemble AI | 1,085±14 |
| 42 | Gradium | 1,084±15 |
| 43 | OpenAI | 1,083±11 |
| 44 | MiniMax | 1,082±12 |
| 45 | Maya Research | 1,081±16 |
| 46 | ElevenLabs | 1,081±11 |
| 47 | 1,074±12 | |
| 48 | Cartesia | 1,071±12 |
| 49 | Mistral | 1,070±14 |
| 50 | Google | 1,070±12 |
| 51 | Fish Audio | 1,066±12 |
| 52 | Speechify | 1,063±13 |
| 53 | OpenAI | 1,062±14 |
| 54 | 1,062±14 | |
| 55 | Amazon | 1,062±12 |
| 56 | Xiaomi | 1,061±14 |
| 57 | StepFun | 1,061±18 |
| 58 | MiniMax | 1,061±11 |
| 59 | Hithink | 1,060±15 |
| 60 | OpenAI | 1,056±22 |
| 61 | Kokoro | 1,055±11 |
| 62 | Hume AI | 1,052±13 |
| 63 | Google | 1,051±12 |
| 64 | Google | 1,050±14 |
| 65 | async | 1,048±12 |
| 66 | Rime | 1,046±14 |
| 67 | Google | 1,044±14 |
| 68 | Fish Audio | 1,044±20 |
| 69 | Maya Research | 1,042±16 |
| 70 | Maya Research | 1,041±12 |
| 71 | Amazon | 1,041±14 |
| 72 | Boson AI | 1,040±15 |
| 73 | 1,037±13 | |
| 74 | Google | 1,036±14 |
| 75 | Cartesia | 1,035±12 |
| 76 | Speechify | 1,033±11 |
| 77 | 1,032±13 | |
| 78 | Microsoft Azure | 1,028±13 |
| 79 | Microsoft Azure | 1,026±19 |
| 80 | MiniMax | 1,024±11 |
| 81 | Smallest.ai | 1,021±13 |
| 82 | Hume AI | 1,021±12 |
| 83 | Xiaomi | 1,016±15 |
| 84 | Resemble AI | 1,012±12 |
| 85 | Fish Audio | 1,010±12 |
| 86 | Rime | 1,003±14 |
| 87 | NVIDIA | 1,002±12 |
| 88 | Zyphra | 1,000±0 |
| 89 | Murf AI | 974.0±12 |
| 90 | LMNT | 968.0±13 |
| 91 | Microsoft Azure | 967.0±14 |
| 92 | StepFun | 959.0±10 |
| 93 | Microsoft Azure | 954.0±14 |
| 94 | OpenVoice | 951.0±14 |
| 95 | NVIDIA | 938.0±15 |
| 96 | Alibaba | 933.0±15 |
| 97 | Neuphonic | 932.0±14 |
| 98 | Alibaba | 915.0±14 |
| 99 | Coqui | 914.0±15 |
| 100 | Google | 902.0±12 |
| 101 | Rime | 888.0±15 |
| 102 | StyleTTS | 884.0±15 |
| 103 | Google | 882.0±12 |
| 104 | Amazon | 879.0±14 |
| 105 | Google | 874.0±12 |
| 106 | Murf AI | 862.0±14 |
| 107 | Noiz | 855.0±17 |
| 108 | MetaVoice | 832.0±18 |
| 109 | Amazon | 809.0±16 |
Artificial Analysis Arena data from Artificial Analysis· updated Aug 1, 2026