数据堂—131小时重口音普通话手机采集语音数据

131小时重口音普通话手机采集语音数据由2568名来自河南、陕西、四川、湖南、福建等省份的中国本土发音人参与录制，男女均衡。131小时重口音普通话手机采集语音为带有较重口音的普通话录制，富有地域特点。录音内容为发音人根据引导问题自由作答的句子

This 131-hour mobile-collected heavy-accent Mandarin speech dataset was recorded by 2568 native Chinese speakers from provinces including Henan, Shaanxi, Sichuan, Hunan, Fujian and other regions, with a balanced gender distribution. The dataset consists of Mandarin speech with prominent regional accents, featuring distinct geographical characteristics. All recordings are spontaneous responses from the speakers in reply to given prompt questions.

搜集汇总

背景与挑战

背景概述

该数据集包含131小时重口音普通话手机采集语音，由2,568名来自河南、陕西、四川等省份的说话者录制，用于中文语音识别模型的测试任务。数据格式为16kHz、16bit的WAV音频，录制环境包括安静室内和正常环境，参与者年龄主要集中在21-30岁，性别分布均衡。