Skip to content
View original post on X: Artificial Analysis· 32/100AI score32/100

HeyGen Voice scores 83.1% on pronunciation robustness benchmark

AISummary

HeyGen Voice scores 83.1% overall on Artificial Analysis's Pronunciation Robustness benchmark, ranking #10 of 29 models.

The benchmark has human reviewers judge whether text-to-speech models pronounce challenging text correctly against pre-agreed accepted pronunciations.

HeyGen Voice places #2 for preserving exact sequences at 83.3%, behind SpaceXAI TTS at 85.7%, and scores 77.0% on expanding shorthand, while Eleven v4 leads that category at 94.1%.

Post on XView on X
Artificial AnalysisVerified on X
@ArtificialAnlys

Part of a thread · earlier post

HeyGen Voice scores 83.1% overall on the Artificial Analysis Pronunciation Robustness benchmark, ranking #10 of 29 models.

Pronunciation Robustness measures whether Text to Speech models correctly pronounce challenging text across four categories, with human reviewers judging each clip against pre-agreed accepted pronunciations.

➤ Preserving Exact Sequences: 83.3%, ranking #2 behind SpaceXAI TTS at 85.7%

➤ Standalone Terms: 90.5%, with Eleven v3 Conversational leading at 95.1%

➤ Contextually Appropriate: 89.8%, with Gemini 3.8 Flash TTS leading at 97.9%

➤ Expanding Shorthand: 77.0%, compared to Qwen-Audio-3.1-TTS-Plus at 80.2%, with Eleven v4 leading at 94.1%

Source: Artificial Analysis · x.comPublished · added here

Topics

Multimodal AIEvals & benchmarksTrends