Controlled Voice Arena LeaderboardArtificial Analysis
Compare Text to Speech (TTS) models using the same 8 cloned voices (4 US, 4 UK).
Range | Model | Elo | Samples | Released | Pricing 1 | |
|---|---|---|---|---|---|---|
| 1 | 1-2 | HeyGen Voice | 1192±1395% interval 1179 to 1205 | 2,807 | Oct 2026 | $30.00 /1M chars |
| 2 | 1-4 | Simba 3.5 | 1189±1795% interval 1172 to 1206 | 1,990 | Nov 2026 | $6.58 /1M chars |
| 3 | 1-5 | HeyGen Voice Turbo | 1178±1595% interval 1163 to 1193 | 2,596 | Oct 2026 | $4.00 /1M chars |
| 4 | 3-5 | Qwen-Audio-3.1-TTS-Plus | 1175±1295% interval 1163 to 1187 | 3,122 | Sept 2026 | $19.31 /1M chars |
| 5 | 3-7 | Qwen-Audio-3.1-TTS-Flash | 1171±1795% interval 1154 to 1188 | 1,763 | - | $19.31 /1M chars |
| 6 | 6-9 | Eleven v4 Turbo | 1156±1295% interval 1144 to 1168 | 3,469 | Sept 2026 | $40.00 /1M chars |
| 7 | 6-9 | Eleven v4 | 1155±1295% interval 1143 to 1167 | 3,352 | Sept 2026 | $80.00 /1M chars |
| 8 | 6-9 | Sonic 3.6 | 1150±1195% interval 1139 to 1161 | 4,217 | Aug 2026 | $49.00 /1M chars |
| 9 | 6-9 | Realtime TTS-2 | 1145±1295% interval 1133 to 1157 | 3,936 | Aug 2026 | $20.83 /1M chars |
| 10 | 10 | Qwen-Audio-3.0-TTS-Plus | 1133±1295% interval 1121 to 1145 | 3,716 | Sept 2026 | $19.31 /1M chars |
| 11 | 11 | Sonic 3.5 | 1118±1095% interval 1108 to 1128 | 6,212 | May 2026 | $49.00 /1M chars |
| 12 | 12-13 | Lightning V3.1 Pro (Oct 2026) | 1101±1495% interval 1087 to 1115 | 2,377 | Oct 2026 | $19.50 /1M chars |
| 13 | 12-15 | AssemblyAI-TTS-1 | 1089±1495% interval 1075 to 1103 | 2,267 | Oct 2026 | $15.00 /1M chars |
| 14 | 13-16 | Simba 3.2 | 1081±1195% interval 1070 to 1092 | 3,761 | Jul 2026 | $6.58 /1M chars |
| 15 | 14-21 | StepAudio 2.5 TTS | 1076±1195% interval 1065 to 1087 | 4,635 | Apr 2026 | $85.00 /1M chars |
| 16 | 14-22 | Eleven v3 | 1075±1195% interval 1064 to 1086 | 5,253 | Feb 2026 | $100.00 /1M chars |
| 17 | 14-23 | 1070±1495% interval 1056 to 1084 | 2,374 | Oct 2026 | $15.00 /1M chars | |
| 18 | 14-23 | 1070±1495% interval 1056 to 1084 | 2,675 | Oct 2026 | N/A | |
| 19 | 14-25 | Voxtral 3 TTS | 1068±1495% interval 1054 to 1082 | 2,570 | Oct 2026 | $13.00 /1M chars |
| 20 | 15-23 | Realtime TTS-2 Flash | 1066±1195% interval 1055 to 1077 | 4,019 | Aug 2026 | $10.42 /1M chars |
| 21 | 15-25 | v3 Conversational | 1066±1295% interval 1054 to 1078 | 3,510 | Aug 2026 | $50.00 /1M chars |
| 22 | 16-25 | SpaceXAI TTS | 1064±1195% interval 1053 to 1075 | 4,210 | Mar 2026 | $15.00 /1M chars |
| 23 | 17-26 | 1059±1295% interval 1047 to 1071 | 2,928 | Oct 2026 | $15.00 /1M chars | |
| 24 | 22-26 | Gradium TTS (Sep 2026) | 1055±1195% interval 1044 to 1066 | 3,059 | Sept 2026 | $47.78 /1M chars |
| 25 | 22-26 | Gemini 3.8 Flash-Lite TTS | 1055±1195% interval 1044 to 1066 | 3,844 | Sept 2026 | $11.03 /1M chars |
| 26 | 22-26 | Gemini 3.8 Flash TTS | 1052±1295% interval 1040 to 1064 | 3,715 | Sept 2026 | $16.49 /1M chars |
| 27 | 27-29 | Simba 3.0 | 1036±1195% interval 1025 to 1047 | 3,920 | Feb 2026 | $6.58 /1M chars |
| 28 | 27-31 | Speech 2.8 HD | 1026±1195% interval 1015 to 1037 | 4,340 | Feb 2026 | $100.00 /1M chars |
| 29 | 27-31 | Speech 2.8 Turbo | 1025±1195% interval 1014 to 1036 | 4,737 | Feb 2026 | $60.00 /1M chars |
| 30 | 28-34 | Gradium TTS (Aug 2026) | 1019±1295% interval 1007 to 1031 | 3,207 | Aug 2026 | $47.78 /1M chars |
| 31 | 28-34 | 1019±1295% interval 1007 to 1031 | 3,256 | Aug 2026 | $34.00 /1M chars | |
| 32 | 30-37 | 1011±1195% interval 1000 to 1022 | 5,285 | Mar 2026 | $15.00 /1M chars | |
| 33 | 30-37 | Fish Audio S2.1 Pro (Oct 2026) | 1008±1395% interval 995 to 1021 | 2,562 | Oct 2026 | $15.00 /1M chars |
| 34 | 30-37 | 1008±1295% interval 996 to 1020 | 3,324 | Jun 2025 | $15.00 /1M chars | |
| 35 | 32-37 | 1005±1295% interval 993 to 1017 | 3,462 | Mar 2026 | $16.00 /1M chars | |
| 36 | 32-37 | Multilingual v2 | 1004±1195% interval 993 to 1015 | 4,368 | Aug 2023 | $100.00 /1M chars |
| 37 | 37 | OpenAudio S1 | 1000AnchorFixed at 1000 as the reference point for this board's Elo scale. | 7,399 | Jun 2025 | $15.00 /1M chars |
| 38 | 37-40 | Flash v2.5 | 990±1195% interval 979 to 1001 | 4,316 | Dec 2024 | $50.00 /1M chars |
| 39 | 38-40 | Fish Audio S1 | 987±1295% interval 975 to 999 | 4,004 | Oct 2025 | $15.00 /1M chars |
| 40 | 38-41 | Turbo v2.5 | 983±1295% interval 971 to 995 | 3,150 | Jul 2024 | $50.00 /1M chars |
| 41 | 40-44 | Kugel 3 | 971±1295% interval 959 to 983 | 4,505 | Jun 2026 | N/A |
| 42 | 41-44 | 963±1395% interval 950 to 976 | 2,964 | Jun 2026 | $15.00 /1M chars | |
| 43 | 41-44 | Sonic 3 | 962±1395% interval 949 to 975 | 3,078 | Oct 2025 | $49.00 /1M chars |
| 44 | 41-44 | Soniox TTS Real-Time v2 | 961±1395% interval 948 to 974 | 2,677 | Aug 2026 | $14.23 /1M chars |
| 45 | 45-48 | 943±1795% interval 926 to 960 | 1,060 | Jul 2026 | N/A | |
| 46 | 45-47 | 940±1395% interval 927 to 953 | 2,983 | May 2025 | $25.00 /1M chars | |
| 47 | 45-48 | Raon SpeechLM | 931±1395% interval 918 to 944 | 2,607 | Apr 2026 | N/A |
| 48 | 47-48 | Noiz TTS | 927±1395% interval 914 to 940 | 2,646 | Nov 2025 | N/A |
| 49 | 49 | Lightning v3.1 | 885±1495% interval 871 to 899 | 2,689 | Mar 2026 | $25.00 /1M chars |
| 50 | 50 | 846±1595% interval 831 to 861 | 2,791 | Nov 2023 | $40.44 /1M chars | |
| 51 | 51 | 820±1695% interval 804 to 836 | 2,398 | Apr 2024 | $8.33 /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.