数据堂—633小时日语语音数据_对话（手机）

日语语音数据_对话（手机），基于30余个常见主题来模拟录制。此数据集标注了文本内容、句时间戳、说话人身份、性别等多种属性，由1000余名来自不同地域和文化背景的日本本土人录制，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

Japanese Mobile-recorded Spoken Dialogue Dataset is simulated and collected based on over 30 common topics. This dataset is annotated with comprehensive attributes including transcript text, sentence-level timestamps, speaker identity, gender and other relevant information. It was recorded by more than 1,000 native Japanese speakers from varied regional and cultural backgrounds, ensuring high data quality and accuracy. It serves as a rich resource for research and applications related to speech recognition, and has been validated by multiple AI companies that it helps models achieve excellent performance when dealing with real-world diversities. We strictly comply with data protection laws and privacy regulations, safeguarding user privacy and legitimate rights and interests throughout the entire process of data collection, storage and utilization. All datasets adhere to the requirements of GDPR, CCPA and PIPL.

搜集汇总

背景与挑战

背景概述

该数据集是数据堂提供的633小时日语对话语音数据，专为语音识别研究和应用测试设计。它基于30多个常见话题，由1000多名日本母语者通过手机录制，标注了文本、时间戳和说话人信息，格式为16kHz单声道WAV，句子准确率达95%。