1,025小时中国重口音普通话语音数据_朗读（手机）

中国重口音普通话语音数据_朗读（手机），基于给定的脚本朗读并模拟录制，涵盖手机语音助手交互、智能家居命令、车载命令词、数字等多种类别，精准匹配智能家居、智能车载等实际应用场景。此数据集标注了文本内容等多种属性，2000余名中国本土发音人参与录制，南方为主，覆盖部分北方重口音省份，男女均衡。语音均采用较重口音的普通话录制，富有地域特点。准确性高，为语音识别相关研究及应用提供了丰富的资源，经多家AI公司验证：有助于模型面对真实世界的多样性时能够表现出色。我们严格遵循数据保护法规和隐私规定，确保数据采集、存储和使用的过程中维护用户的隐私和合法权益，所有数据均遵循GDPR, CCPA, PIPL。

Mandarin Speech Dataset with Heavy Regional Chinese Accents (Scripted Read Speech, Mobile Collection). This dataset comprises scripted read speech recordings collected via mobile devices, covering multiple categories including mobile voice assistant interactions, smart home commands, in-vehicle command terms, numerals and other scenarios, which precisely aligns with real-world application scenarios such as smart home and intelligent vehicle systems. The dataset is annotated with multiple attributes including text content, and was recorded by over 2,000 native Chinese speakers, with the majority coming from southern China and covering some northern provinces with heavy accents, maintaining a balanced gender ratio. All speech samples are recorded in Mandarin with prominent regional accents, featuring distinct geographic characteristics. With high data quality, this dataset provides abundant resources for speech recognition-related research and applications, and has been validated by multiple AI companies that it enables models to perform excellently when confronting real-world diversity. We strictly comply with data protection laws and privacy regulations, ensuring that user privacy and legitimate rights and interests are fully protected throughout the entire process of data collection, storage and usage. All data adheres to GDPR, CCPA and PIPL.

搜集汇总

背景与挑战

背景概述

该数据集是一个1,025小时的中国重口音普通话语音数据集，通过手机录制，涵盖智能家居、车载命令等多种应用场景，由2,444名发音人参与，口音较重且地域多样，标注准确率高，适用于语音识别研究。数据集注重隐私保护，符合国际数据法规标准。