20小时四川话女声合成库

20小时四川话女声合成库，由四川成都发音人录制。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。

20-hour Sichuan dialect female speech synthesis corpus, recorded by a female speaker from Chengdu, Sichuan Province. The corpus features well-balanced phoneme coverage, was annotated by professional phoneticians, and is precisely tailored to the R&D requirements of speech synthesis technologies.

搜集汇总

背景与挑战

背景概述

这是一个专为语音合成设计的四川方言女声数据集，包含20小时由四川成都女性专业声优在专业录音棚录制的音频，采用48kHz/24bit的wav格式。数据集具有音素覆盖均衡、音字和韵律标注精细的特点，旨在精准支持四川话语音合成模型的研发需求。