HeyGen Voice beats ElevenLabs and Cartesia in blind ambient-fit tests
HeyGen · x · 2026-10-15
HeyGen shared blind-test results for its Voice model's ambient fit: one voice placed in 50 settings (a cathedral, a factory floor, a car's front seat), with reviewers judging whether the sound matches the picture without labels. HeyGen Voice won 73% vs Cartesia, 70% vs ElevenLabs v4, and 65% vs Seedance 2.5. The company claims the model ranks #1 on Artificial Analysis and is the first voice model built for video, performing scripts in the user's own voice, pacing and expression while adapting to scene acoustics.
Related event: HeyGen launches Voice cloning model, claims blind-test wins over ElevenLabs(6 posts)→
More from Models
- Silence Is Billable on Voice Models: Even Dead Air Costs Money — sophiamyang · 2026-10-15
- NCM: new text generation paradigm up to 31x faster than diffusion LMs with better quality — simonguozirui · 2026-10-15
- Gemma 4 26B priced at $0.07/M input tokens, ~40% below comparable providers — testingcatalog · 2026-10-15
- Epoch AI: GPT-6 Astra tops frontier models in Slay the Spire 2 but still loses to top humans — sandersted · 2026-10-15
- Google execs cite early signs of recursive self-improvement behind 3-4 week Gemini cadence — danielrock · 2026-10-15
- MoralityBench tests 16 LLMs with validated moral psychology surveys; only 3 exceed 90% consistency — NakedPlato · 2026-10-15