数据堂—2人日语平均音色合成库

本套数据2人日语平均音色合成库由日本本土发音人参与录制，发音纯正，包含新闻和口语化通用语料，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 2-speaker Japanese average voice synthesis corpus was recorded by native Japanese speakers with authentic pronunciation. It encompasses both news-style and colloquial general-purpose speech materials, featuring balanced phoneme coverage across all utterances. Annotated by professional phoneticians, this corpus accurately meets the R&D requirements of speech synthesis.

搜集汇总

背景与挑战

背景概述

该数据集是一个日语平均音色合成库，包含新闻和口语材料，由两位专业配音演员录制，每人10小时，专为语音合成研发设计。数据格式为48kHz、24bit的WAV文件，适用于商业用途。