Researchers have introduced the CN-NewsTTS Bench, a new benchmark designed to evaluate the pronunciation accuracy of Chinese news Text-to-Speech (TTS) systems. This benchmark specifically targets complex written forms like scores, hyphenated names, and mixed Chinese-Latin-digit expressions, which are common in news content and can be mispronounced by TTS systems. The benchmark includes development and test sets, auto-evaluable targets, and transcripts from an Automatic Speech Recognition (ASR) ensemble, with initial results showing the best performing system achieving 0.879 accuracy, while others lag significantly below 0.60. AI
IMPACT This benchmark aims to improve the naturalness and accuracy of Chinese news TTS, potentially leading to better voice assistants and audio content generation.
RANK_REASON The cluster describes a new benchmark for evaluating TTS systems, which falls under research.
Read on Hugging Face Daily Papers →
AI-generated summary · Google Gemini · from 3 sources. How we write summaries →
