数据堂—13.3小时中文女声情感合成库

13.3小时中文女声情感合成库，由中文母语发音人录制，六种情感文本，语料音素覆盖均衡，专业语音学家参与标注，精准匹配语音合成的研发需求。

This is a 13.3-hour Mandarin female emotional speech synthesis corpus recorded by native Mandarin speakers. It features text prompts for six distinct emotional categories, with balanced phonetic coverage across the entire dataset. Annotated by professional phoneticians, this corpus is precisely tailored to meet the research and development needs of speech synthesis systems.

搜集汇总

背景与挑战

背景概述

该数据集是一个13.3小时的中文女声情感语音合成库，包含六种情感（快乐、愤怒、悲伤、惊讶、恐惧、厌恶）的录音，由20-30岁女性录制，音频格式为48kHz、16bit、wav单声道。它专用于中文女声情感语音合成模型的测试任务，由数据堂拥有商业版权。