数据堂—502小时中国人说英语手机采集语音数据

502小时中国人说英语手机采集语音数据是由1279名中国发音人参与录制，覆盖国内各大方言区，符合中国人说英语的特定口音。录音文本涵盖出国口语、演讲、人机交互等多类别，内容丰富，领域广泛，音素均衡。本套中国人说英语手机采集语音数据可用于改善语音识别系统对中国人说英语的识别效果。

This 502-hour mobile-collected spoken English speech dataset was recorded by 1,279 Chinese speakers from major dialect regions across China, and aligns with the characteristic English accent of Chinese speakers. The recorded texts cover multiple categories including daily oral English for overseas travel, public speeches, human-computer interaction scenarios and more, boasting rich content, wide domain coverage and balanced phoneme distribution. This dataset can be utilized to enhance the recognition performance of speech recognition systems for English spoken by Chinese speakers.

搜集汇总

背景与挑战

背景概述

该数据集包含502小时中国人说英语的手机录音，由1,279名说话者录制，覆盖中国主要方言区，内容涵盖日常对话、旅行、演讲和交互等多个类别。它专用于测试英语语音识别模型，旨在提升系统对中国人说英语的识别性能。