1,027人唤醒词语音数据_朗读（麦克风）【数据堂】

唤醒词语音数据_朗读（麦克风），基于给定的唤醒词朗读并录制，覆盖慢速、匀速、快速三种语速。此数据集标注了文本内容、口音、出生地、性别等多种属性，由1027名来自中国8个不同省份的人员录制，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

Wake Word Speech Data - Read Speech (Microphone): Recorded by reading aloud a given wake word, this dataset covers three speech speeds: slow, normal, and fast. It is annotated with multiple attributes including text content, accent, place of birth, gender, and more. Recorded by 1027 participants from 8 distinct provinces across China, this dataset features high accuracy and provides abundant resources for speech recognition-related research and applications. It has been validated by multiple AI companies to help models perform excellently when confronted with real-world diversities. We strictly comply with data protection regulations and privacy rules, ensuring that user privacy and legitimate rights and interests are fully protected during data collection, storage, and utilization, and all datasets adhere to the requirements of GDPR, CCPA, and PIPL.

搜集汇总

背景与挑战

背景概述

该数据集包含1027名中国人员朗读20个常用唤醒词的语音数据，覆盖慢速、匀速和快速三种语速，并标注了文本内容、口音、出生地、性别等多种属性。数据采用48kHz未压缩wav格式，在专业录音棚中使用高保真麦克风录制，专为语音识别研究设计，但仅限企业合作购买使用。