中英混读手机采集语音数据【数据堂】

1,535小时中英混读手机采集语音数据由3972名中国本土人员参与录制，口音覆盖七大方言区。录音文本均为中英混合句子，涵盖通用场景及人机交互场景，内容丰富，转写精准。可用于改善语音识别系统对中英混读语音的识别效果。

This dataset contains 1,535 hours of Mandarin-English code-switching speech data collected via mobile phones, recorded by 3,972 native Chinese speakers. The accents of the recorded speech cover seven major Chinese dialect regions. All transcribed texts are Mandarin-English mixed sentences, covering both general daily scenarios and human-computer interaction scenarios, with rich content and accurate transcriptions. It can be utilized to improve the recognition performance of speech recognition systems for Mandarin-English code-switching speech.

搜集汇总

背景与挑战

背景概述

该数据集包含1,535小时的中英混读手机采集语音，由3,972名中国本土人员录制，覆盖七大方言区，录音文本为中英混合句子，涵盖通用和人机交互场景，转写精准，旨在提升语音识别系统对中英混读语音的识别能力。此数据集为商业数据，仅面向企业合作购买。