数据堂6.78小时中文女声模仿童声合成库

中文女声模仿童声合成库为女声成人模仿儿童的语音数据，共计6599句，6.78小时，由中文母语发音人录制，发音纯正，音色甜美，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 Chinese female-to-child voice imitation speech synthesis dataset, which consists of speech data where adult female native Chinese speakers imitate children's voices. It totals 6599 utterances and 6.78 hours of audio, recorded by native Chinese speakers with authentic pronunciation and sweet timbre. The corpus has a balanced coverage of phonemes, and was annotated by professional phoneticians, perfectly meeting the R&D requirements of speech synthesis.

搜集汇总

背景与挑战

背景概述

该数据集是一个专门用于中文语音识别模型测试的语音合成库，包含6.78小时的中文女声模仿童声音频，共计6599句，具有音素覆盖均衡和专业标注的特点。数据以48kHz、24bit的wav格式录制，在专业录音棚环境中采集，适用于语音合成研发。