Eleven v4 tops TTS Arena at Elo 1,319 and record 91.7% pronunciation score
ArtificialAnlys · x · 2026-09-28
Artificial Analysis evaluated ElevenLabs' Eleven v4:
- #1 on Provider Voice TTS Arena with Elo 1,319 (1,674 appearances), ahead of Cartesia Sonic 3.6 (1,276) and Google Gemini 3.8 Flash TTS (1,267); first place in all four categories.
- #2 on Controlled Voice (Elo 1,157), just behind Alibaba's Qwen-Audio-3.1-TTS-Plus (1,178) and far above Eleven v3 (1,073).
- Pronunciation Robustness 91.7%, the highest ever measured, up from v3's 85.6%; above 93% in three of four subcategories.
- Language support expands from 70+ to 90+.
- Cost: $80/1M characters vs $49 for Sonic 3.6 and $16.49 for Gemini 3.8 Flash TTS; speed up to 73.4 chars/sec from v3's 42.5.
More from Multimodal
- Eleven v4 tops Artificial Analysis rankings; ElevenLabs gives away 11k credits for 24 hours — charis_ai · 2026-09-28
- Horse gaits one-shot with Claude Opus 5.5: impressive coat shine and musculature — shekitup · 2026-09-28
- Ad Buyer Ditches Image Models, Uses Seedance 2.5 for Realistic AI UGC Characters — churchkey · 2026-09-28
- One-word Midjourney prompt: 'Pulchritudinous' at --ar 5:4 --v 8.2 — tisch_eins · 2026-09-28
- Prompt Template Makes Qwen Image 2.1 Design Like Closed-Source Models — wjc_5 · 2026-09-28
- A ChatGPT prompt turns GPT Image into an editorial fashion photographer — umesh_ai · 2026-09-28