TTS leaderboard shakeup: accent alone moves Elo, HeyGen Voice tops overall board
ArtificialAnlys · x · 2026-11-07
Artificial Analysis ran a controlled Voice Arena across 49 TTS models and found accent alone can shift results: Qwen-Audio-3.1-TTS-Plus leads in US English, while HeyGen Voice performs better in UK English and tops the overall leaderboard — with only the target accent changed.
Related event: Accent Matters: 49 TTS Models Tested in Controlled Voice Arena(4 posts)→
More from Models
- Hobbyist's 'Weight Space Programming' on Qwen 3.8 cuts token usage up to 58% with no accuracy loss — Xananique · 2026-11-07
- Inkling 2 Small pretrained on 8.4T tokens in 68 hours, natively in NVFP4 — teortaxesTex · 2026-11-07
- InstantHMR open-sourced: real-time 3D human mesh recovery in 5.5ms per person on phone NPU — PurzBeats · 2026-11-07
- Bittensor week: subnets hit model Pareto frontier, GPU rentals hit $2.32M in October — markjeffrey · 2026-11-07
- Local ripple-65B scores 38/49/27 on agentic coding: errors encountered, not draft quality, decided the grade — Ok-Brush8377 · 2026-11-07
- 'o1 was AI's biggest moment' — Epoch chart shows post-o1 gains turned incremental — flowersslop · 2026-11-07