数据堂—183小时乌尔都语语音数据_对话（手机）

乌尔都语语音数据_对话（手机），基于约三十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240名来自不同地域和文化背景的母语人录制，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。

Urdu Speech Dialogue Dataset (Mobile-Collected) is simulated and recorded based on approximately 30 common topics. This dataset is annotated with multiple attributes including transcript text, sentence-level timestamps, speaker identity, gender and other relevant information. It was recorded by 240 native speakers from diverse regional and cultural backgrounds, with high overall data accuracy. The dataset provides abundant resources for speech recognition-related research and applications, and has been validated by multiple AI companies that it helps models deliver excellent performance when facing real-world diversities. We strictly comply with data protection regulations and privacy rules to safeguard users' privacy and legitimate rights and interests throughout the entire process of data collection, storage and usage. All collected data adheres to the compliance 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183小时的乌尔都语手机对话语音，基于30个常见话题由240名母语者录制，标注了文本内容、时间戳和说话人信息。音频格式为16kHz的wav文件，字准确率达95%，适用于语音识别研究，遵循Apache 2.0许可但仅限商业使用。