849小时中国普通话语音数据_交互场景（手机）【数据堂】

中国普通话语音数据_交互场景（手机），在真实的家居场景下进行录制，录音内容以家居指令、功能性助手和唤醒词为主，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

Chinese Mandarin Speech Dataset: Interactive Scenarios (Mobile Phones)
This dataset was recorded in real-world home scenarios, with the recorded content primarily comprising home commands, functional assistant utterances and wake words. Boasting high accuracy, it offers rich resources for research and applications related to speech recognition. Validated by multiple AI companies, this dataset assists models in delivering outstanding performance when confronting the diversity of real-world scenarios. We strictly comply with data protection regulations and privacy rules to ensure that user privacy and legitimate rights and interests are protected throughout the processes of data collection, storage and usage. All data complies with GDPR, CCPA and PIPL.

搜集汇总

背景与挑战

背景概述

该数据集包含849小时中国普通话语音数据，专为手机交互场景设计，在真实家居环境中录制，内容涵盖家居指令、功能性助手和唤醒词等。语音格式为48kHz、16bit的未压缩wav单声道音频，由998名说话人参与录制，句准确率达98%，适用于语音识别研究，但仅限企业合作购买。