Provider Voice Arena LeaderboardArtificial Analysis
Compare Text to Speech (TTS) models using each provider's own native voices.
Range | Model | Elo | Samples | Arena Voices | Released | Pricing 1 | |
|---|---|---|---|---|---|---|---|
| 1 | 1-2 |  Eleven v4 | 1334±1595% interval 1319 to 1349 | 2,608 | Sept 2026 | $80.0 /1M chars | |
| 2 | 1-2 | Eleven v4 Turbo | 1331±1695% interval 1315 to 1347 | 2,311 | Sept 2026 | $40.0 /1M chars | |
| 3 | 3-4 |  Qwen-Audio-3.1-TTS-Plus | 1301±1595% interval 1286 to 1316 | 2,325 | Sept 2026 | $19.3 /1M chars | |
| 4 | 3-5 |  Sonic 3.6 | 1293±1395% interval 1280 to 1306 | 3,176 | Aug 2026 | $49.0 /1M chars | |
| 5 | 4-7 |  Gemini 3.8 Flash TTS | 1282±1395% interval 1269 to 1295 | 3,442 | Sept 2026 | $16.5 /1M chars | |
| 6 | 5-10 |  Fish Audio Drama 3  | 1272±1895% interval 1254 to 1290 | 1,561 | Oct 2026 | $15.0 /1M chars | |
| 7 | 5-10 |  Lightning V3.1 Pro (Oct 2026) | 1269±1595% interval 1254 to 1284 | 1,874 | Oct 2026 | $19.5 /1M chars | |
| 8 | 6-10 | Qwen-Audio-3.0-TTS-Plus | 1266±1395% interval 1253 to 1279 | 2,844 | Sept 2026 | $19.3 /1M chars | |
| 9 | 6-10 |  Realtime TTS-2 | 1266±1495% interval 1252 to 1280 | 2,564 | Aug 2026 | $20.8 /1M chars | |
| 10 | 8-11 | Gemini 3.8 Flash-Lite TTS | 1254±1395% interval 1241 to 1267 | 3,432 | Sept 2026 | $11.0 /1M chars | |
| 11 | 10-12 |  Simba 3.2 | 1247±1295% interval 1235 to 1259 | 3,804 | Jul 2026 | $6.6 /1M chars | |
| 12 | 11-13 | Fish Audio Drama 3 Preview | 1239±1495% interval 1225 to 1253 | 2,593 | Oct 2026 | $15.0 /1M chars | |
| 13 | 12-16 |  Breeze TTS 2 | 1226±1395% interval 1213 to 1239 | 2,735 | Aug 2026 | $34.0 /1M chars | |
| 14 | 13-19 |  AssemblyAI-TTS-1 | 1221±1695% interval 1205 to 1237 | 1,707 | Oct 2026 | $15.0 /1M chars | |
| 15 | 13-19 |  Voxtral 3 TTS | 1218±1495% interval 1204 to 1232 | 2,076 | Oct 2026 | $13.0 /1M chars | |
| 16 | 13-20 | Realtime TTS-2 Flash | 1215±1295% interval 1203 to 1227 | 3,050 | Aug 2026 | $10.4 /1M chars | |
| 17 | 14-20 | Gemini 3.1 Flash TTS | 1213±1195% interval 1202 to 1224 | 4,643 | Apr 2026 | $18.3 /1M chars | |
| 18 | 14-20 | v3 Conversational | 1212±1295% interval 1200 to 1224 | 3,189 | Aug 2026 | $50.0 /1M chars | |
| 19 | 14-20 |  StepAudio 2.5 TTS (Aug 2026) | 1210±1395% interval 1197 to 1223 | 2,663 | Aug 2026 | $85.0 /1M chars | |
| 20 | 17-20 |  Luna TTS | 1204±1195% interval 1193 to 1215 | 3,899 | Jun 2026 | $15.0 /1M chars | |
| 21 | 21-24 | Sonic 3.5 | 1193±1195% interval 1182 to 1204 | 4,167 | May 2026 | $49.0 /1M chars | |
| 22 | 21-25 |  Ming-3.0-tts | 1189±1395% interval 1176 to 1202 | 2,484 | Oct 2026 | N/A | |
| 23 | 21-25 | Eleven v3 | 1187±1095% interval 1177 to 1197 | 5,428 | Feb 2026 | $100.0 /1M chars | |
| 24 | 21-27 | Fish Audio S2.1 Pro (Oct 2026) | 1183±1395% interval 1170 to 1196 | 2,475 | Oct 2026 | $15.0 /1M chars | |
| 25 | 23-27 |  Soniox TTS Real-Time v2 | 1179±1095% interval 1169 to 1189 | 5,797 | Aug 2026 | $14.2 /1M chars | |
| 26 | 24-28 |  Speech 2.8 HD | 1173±1095% interval 1163 to 1183 | 5,545 | Feb 2026 | $100.0 /1M chars | |
| 27 | 24-28 |  Gradium TTS (Sep 2026) | 1172±1395% interval 1159 to 1185 | 2,589 | Sept 2026 | $47.8 /1M chars | |
| 28 | 25-30 | Gradium TTS (Oct 2026) | 1166±1495% interval 1152 to 1180 | 1,995 | Sept 2026 | $47.8 /1M chars | |
| 29 | 28-32 |  SpaceXAI TTS | 1157±1395% interval 1144 to 1170 | 2,186 | Mar 2026 | $15.0 /1M chars | |
| 30 | 28-32 |  Falcon 2 | 1156±1295% interval 1144 to 1168 | 2,807 | Jul 2026 | $10.0 /1M chars | |
| 31 | 29-33 | Gradium TTS (Aug 2026) | 1151±1295% interval 1139 to 1163 | 2,833 | Aug 2026 | $47.2 /1M chars | |
| 32 | 29-33 | Speech 2.8 Turbo | 1151±1095% interval 1141 to 1161 | 5,301 | Feb 2026 | $60.0 /1M chars | |
| 33 | 31-35 |  Azure HD 2.5 | 1143±1195% interval 1132 to 1154 | 3,270 | Mar 2026 | $22.0 /1M chars | |
| 34 | 33-35 |  Async Flash v1.5 | 1139±1195% interval 1128 to 1150 | 2,971 | May 2026 | $10.1 /1M chars | |
| 35 | 33-37 | Step TTS 2 (Mar 2026) | 1136±1195% interval 1125 to 1147 | 3,051 | Feb 2026 | $40.0 /1M chars | |
| 36 | 35-38 | Async Pro v1.0 | 1127±1195% interval 1116 to 1138 | 2,829 | May 2026 | $20.2 /1M chars | |
| 37 | 35-39 | Speech 2.6 HD | 1125±1195% interval 1114 to 1136 | 4,539 | Oct 2025 | $100.0 /1M chars | |
| 38 | 36-40 | Speech 2.6 Turbo | 1120±1095% interval 1110 to 1130 | 4,783 | Oct 2025 | $60.0 /1M chars | |
| 39 | 37-45 | Simba 3.0 | 1115±1195% interval 1104 to 1126 | 3,257 | Feb 2026 | $6.6 /1M chars | |
| 40 | 38-46 | Phonon 31M (English) | 1111±1495% interval 1097 to 1125 | 1,821 | Oct 2026 | $0.0 /1M chars | |
| 41 | 38-46 |  Chatterbox HD | 1108±1295% interval 1096 to 1120 | 2,860 | — | May 2025 | $40.0 /1M chars |
| 42 | 39-46 | Fish Audio S2 Pro | 1108±1195% interval 1097 to 1119 | 2,907 | Mar 2026 | $15.0 /1M chars | |
| 43 | 38-47 |  Coda v2 | 1108±1495% interval 1094 to 1122 | 1,840 | Oct 2026 | $50.0 /1M chars | |
| 44 | 39-46 | Speech-02-HD | 1106±1195% interval 1095 to 1117 | 4,844 | Mar 2025 | $100.0 /1M chars | |
| 45 | 39-47 |  TTS-1 HD | 1106±1195% interval 1095 to 1117 | 3,976 | Nov 2023 | $30.0 /1M chars | |
| 46 | 41-52 | Step Audio EditX (Mar 2026) | 1097±1295% interval 1085 to 1109 | 2,518 | Feb 2026 | N/A | |
| 47 | 44-54 |  Speech 2.6 | 1095±1295% interval 1083 to 1107 | 2,716 | Jul 2026 | N/A | |
| 48 | 46-54 | Turbo v2.5 | 1092±1095% interval 1082 to 1102 | 7,216 | Jul 2024 | $50.0 /1M chars | |
| 49 | 46-54 | Multilingual v2 | 1092±1095% interval 1082 to 1102 | 7,647 | Aug 2023 | $100.0 /1M chars | |
| 50 | 46-54 | Speech-02-Turbo | 1090±1195% interval 1079 to 1101 | 4,161 | Mar 2025 | $60.0 /1M chars | |
| 51 | 46-54 | Voxtral TTS | 1089±1295% interval 1077 to 1101 | 2,634 | Mar 2026 | $16.0 /1M chars | |
| 52 | 47-54 | TTS-1 | 1086±1095% interval 1076 to 1096 | 6,450 | Nov 2023 | $15.0 /1M chars | |
| 53 | 47-55 | OpenAudio S1 | 1085±1195% interval 1074 to 1096 | 4,787 | Jun 2025 | $15.0 /1M chars | |
| 54 | 48-57 | Gemini 2.5 Flash Lite TTS | 1083±1195% interval 1072 to 1094 | 3,458 | Dec 2025 | $9.2 /1M chars | |
| 55 | 54-60 | Flash v2.5 | 1075±1095% interval 1065 to 1085 | 5,962 | Dec 2024 | $50.0 /1M chars | |
| 56 | 54-62 | T2A-01-HD | 1072±1195% interval 1061 to 1083 | 5,192 | Jan 2025 | $50.0 /1M chars | |
| 57 | 54-64 | GPT-Realtime-2 | 1072±1295% interval 1060 to 1084 | 2,459 | May 2026 | $191.6 /1M chars | |
| 58 | 55-62 | Studio | 1072±1195% interval 1061 to 1083 | 4,735 | Feb 2023 | $160.0 /1M chars | |
| 59 | 55-66 | Journey | 1070±1295% interval 1058 to 1082 | 2,458 | Dec 2023 | $160.0 /1M chars | |
| 60 | 55-67 | Gemini 2.5 Flash TTS (Dec 2025) | 1068±1295% interval 1056 to 1080 | 3,057 | Dec 2025 | $9.2 /1M chars | |
| 61 | 55-70 |  Magpie-Multilingual 357M (Feb 2026) | 1064±1295% interval 1052 to 1076 | 2,411 | Feb 2026 | N/A | |
| 62 | 56-70 |  Polly Generative | 1063±1195% interval 1052 to 1074 | 3,919 | May 2024 | $30.0 /1M chars | |
| 63 | 59-70 |  Kokoro 82M v1.0 | 1061±1195% interval 1050 to 1072 | 5,545 | Jan 2025 | $0.7 /1M chars | |
| 64 | 59-70 | Chirp 3: HD | 1060±1195% interval 1049 to 1071 | 3,547 | Mar 2025 | $30.0 /1M chars | |
| 65 | 57-72 |  Maya 2 Flash | 1059±1395% interval 1046 to 1072 | 2,326 | Jul 2026 | N/A | |
| 66 | 59-70 |  Octave 2 | 1059±1295% interval 1047 to 1071 | 3,305 | Oct 2025 | $87.5 /1M chars | |
| 67 | 60-72 | Sonic 3 | 1058±1195% interval 1047 to 1069 | 3,664 | Oct 2025 | $49.0 /1M chars | |
| 68 | 61-74 | Coda | 1052±1295% interval 1040 to 1064 | 2,593 | May 2026 | $50.0 /1M chars | |
| 69 | 61-74 |  MiMo-V2.5-TTS | 1052±1295% interval 1040 to 1064 | 2,234 | - | N/A | |
| 70 | 61-75 |  Flux TTS | 1052±1395% interval 1039 to 1065 | 2,354 | Aug 2026 | $45.0 /1M chars | |
| 71 | 65-78 | Maya 2 Global | 1047±1395% interval 1034 to 1060 | 2,129 | Jul 2026 | N/A | |
| 72 | 67-77 | Async Flash v1.0 | 1047±1195% interval 1036 to 1058 | 4,167 | Jul 2025 | $10.1 /1M chars | |
| 73 | 68-78 | Maya1 | 1045±1295% interval 1033 to 1057 | 3,433 | Jun 2025 | N/A | |
| 74 | 63-82 | OpenAudio S1 Mini | 1043±2095% interval 1023 to 1063 | 1,682 | Jun 2025 | $15.0 /1M chars | |
| 75 | 71-80 | Gemini 2.5 Pro (Dec 2025) | 1040±1295% interval 1028 to 1052 | 2,554 | Dec 2025 | $18.3 /1M chars | |
| 76 | 71-81 | Octave TTS | 1038±1195% interval 1027 to 1049 | 4,793 | Feb 2025 | $87.5 /1M chars | |
| 77 | 71-82 |  Higgs Audio V3 TTS | 1037±1395% interval 1024 to 1050 | 2,318 | Jun 2026 | $15.0 /1M chars | |
| 78 | 73-82 |  Bland Speech v3 | 1035±1295% interval 1023 to 1047 | 2,692 | Aug 2026 | $40.0 /1M chars | |
| 79 | 70-84 | Azure Neural | 1032±2095% interval 1012 to 1052 | 1,156 | Sept 2018 | $15.0 /1M chars | |
| 80 | 75-84 | MAI-Voice-1 | 1030±1295% interval 1018 to 1042 | 2,423 | Apr 2026 | N/A | |
| 81 | 76-84 | T2A-01-Turbo | 1027±1195% interval 1016 to 1038 | 5,075 | Jan 2025 | $30.0 /1M chars | |
| 82 | 77-84 | Polly Long-Form | 1025±1395% interval 1012 to 1038 | 2,036 | Nov 2023 | $100.0 /1M chars | |
| 83 | 79-84 | Chatterbox  | 1022±1195% interval 1011 to 1033 | 4,785 | May 2025 | $25.0 /1M chars | |
| 84 | 79-84 | Lightning v3.1 | 1022±1295% interval 1010 to 1034 | 2,440 | Mar 2026 | $25.0 /1M chars | |
| 85 | 85-86 |  Raon SpeechLM | 1006±1395% interval 993 to 1019 | 2,332 | Apr 2026 | N/A | |
| 86 | 86 |  Zonos-v0.1 | 1000AnchorFixed at 1000 as the reference point for this board's Elo scale. | 4,932 | Feb 2025 | $20.0 /1M chars | |
| 87 | 87 | Murf Speech Gen 2 | 985±1295% interval 973 to 997 | 5,078 | Jul 2024 | $100.0 /1M chars | |
| 88 | 88-89 |  OpenVoice v2 | 956±1395% interval 943 to 969 | 3,128 | — | Apr 2024 | $8.3 /1M chars |
| 89 | 88-91 | VibeVoice 1.5B | 953±1495% interval 939 to 967 | 2,198 | Aug 2025 | N/A | |
| 90 | 89-92 |  Neuphonic TTS | 940±1495% interval 926 to 954 | 2,524 | Oct 2024 | $20.8 /1M chars | |
| 91 | 89-92 | Qwen3 TTS Flash | 939±1495% interval 925 to 953 | 2,240 | Sept 2025 | $11.0 /1M chars | |
| 92 | 90-94 | Qwen3 TTS | 929±1395% interval 916 to 942 | 3,178 | Jan 2026 | $13.8 /1M chars | |
| 93 | 92-95 |  Kugel 3 | 924±1395% interval 911 to 937 | 4,517 | Jun 2026 | N/A | |
| 94 | 93-95 | WaveNet | 917±1295% interval 905 to 929 | 6,442 | Sept 2016 | $4.0 /1M chars | |
| 95 | 92-96 |  XTTS v2 | 915±1595% interval 900 to 930 | 2,740 | — | Nov 2023 | $40.4 /1M chars |
| 96 | 94-99 | Mist V2 | 902±1595% interval 887 to 917 | 3,089 | Mar 2025 | $31.1 /1M chars | |
| 97 | 96-100 | Polly Neural | 893±1495% interval 879 to 907 | 3,000 | Jul 2019 | $16.0 /1M chars | |
| 98 | 96-100 | Neural2 | 891±1295% interval 879 to 903 | 6,719 | Jun 2022 | $16.0 /1M chars | |
| 99 | 96-100 |  StyleTTS 2 | 890±1595% interval 875 to 905 | 2,721 | — | Jun 2023 | $2.8 /1M chars |
| 100 | 97-100 | Standard | 886±1295% interval 874 to 898 | 6,796 | Aug 2018 | $4.0 /1M chars | |
| 101 | 101 |  Noiz TTS | 866±1595% interval 851 to 881 | 2,484 | Nov 2025 | N/A | |
| 102 | 102-103 |  MetaVoice v1 | 846±1895% interval 828 to 864 | 1,690 | — | Feb 2024 | N/A |
| 103 | 103 | Polly Standard | 830±1595% interval 815 to 845 | 3,622 | 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
2 Open Weights includes models that can be self-hosted using downloadable model weights or a provider-supplied inference runtime. Open Weights Commercial includes models whose published license permits commercial use, including use subject to license conditions (e.g. an NVIDIA AI Enterprise subscription), and excludes models that require a separately negotiated commercial agreement.
Frequently Asked Questions
The top Text to Speech models by Elo rating are: 1. Eleven v4 (Elo 1334), 2. Eleven v4 Turbo (Elo 1331), 3. Qwen-Audio-3.1-TTS-Plus (Elo 1301), 4. Sonic 3.6 (Elo 1293), and 5. Gemini 3.8 Flash TTS (Elo 1282). 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 1061. 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 Drama 3 (Elo 1272), 2. Fish Audio Drama 3 Preview (Elo 1239), 3. Breeze TTS 2 (Elo 1226), 4. Ming-3.0-tts (Elo 1189), and 5. Phonon 31M (English) (Elo 1111).