数据堂—299小时美国儿童语音数据_朗读（手机）

美国儿童语音数据_朗读（手机）基于给定的脚本朗读并录制，涵盖儿童读物、教科书等多类别，内容丰富，符合儿童语言使用习惯。此数据集标注了文本内容等多种属性，由200余名来自不同地域和文化背景的美国儿童参与录制，口音正宗，文本经过人工校对，准确率高，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

US Children's Speech Dataset (Reading Aloud via Mobile Phones): This dataset is collected by having participants read aloud pre-provided scripts, covering diverse categories including children's books and textbooks, with rich content that aligns with natural language usage habits of American children. Multiple attributes such as the original transcribed text are annotated for this dataset. It was recorded by over 200 American children from different regional and cultural backgrounds, featuring authentic accents. The transcribed texts have undergone manual proofreading to ensure high accuracy. This dataset provides abundant resources for speech recognition-related research and applications. Verified by multiple AI companies, it helps models deliver excellent performance when coping with the diversity of real-world scenarios. We strictly comply with data protection regulations and privacy rules, safeguarding user privacy and legitimate rights and interests throughout the entire process of data collection, storage and usage. All data adheres 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299小时的美国儿童朗读语音数据，专为语音识别研究和应用设计。数据由200多名5-12岁儿童录制，涵盖儿童文学和教科书等内容，录音格式为16kHz WAV，并经过人工校对确保高准确性。