Cartesia's Sonic 3.6 tops both speech arena leaderboards, nearly 3x faster than Eleven v3
ArtificialAnlys · x · 2026-08-18
Cartesia's latest TTS model Sonic 3.6 takes #1 on both the Provider Voice and Controlled Voice Artificial Analysis Speech Arena leaderboards, surpassing Speechify's Simba 3.2 and Alibaba's Qwen-Audio-3.0-TTS-Plus.
- Rankings: #1 on Controlled Voice with Elo 1144 (1,339 appearances), ahead of its own Sonic 3.5 (1099) and ElevenLabs' Eleven v3 (1060); #1 on Provider Voice at 1286, beating Simba 3.2 (1238) and Qwen (1237)
- Pricing: $49 per 1M characters — 5x Simba 3.2 ($10), nearly 2x Qwen ($27.59), but half of Eleven v3 ($100)
- Speed: 136.1 chars/sec generation vs 46.7 for Eleven v3 and 24.2 for Google's Gemini 3.1 Flash TTS
- Supports 40+ languages, with voters citing natural speech and adaptive pacing
Related event: Cartesia Releases Sonic-3.6, Topping Speech Leaderboards(4 posts)→
More from Multimodal
- MLP-Splatting (ECCV 2026) code released: MLPs as light-field primitives instead of Gaussians — AjdDavison · 2026-10-16
- Digital human Fae codes her own facial expressions frame-by-frame with Claude Opus 5.5 and FACS — repligate · 2026-10-16
- Meta's Omnilingual ASR, covering 1,600 languages, named a TIME Best Invention of 2026 — AIatMeta · 2026-10-16
- ATAS: Zhejiang-led team builds world-state memory framework for consistent agentic video generation — CSProfKGD · 2026-10-16
- Flux 3 Klein's Default Templates Are Broken Without Its Prompt Upscaler — StrictMove5839 · 2026-10-16
- Kroma Touted as the Most Well-Rounded Image Model Yet on Civitai — Different_Fix_2217 · 2026-10-16