PharynxAI/IndicVoices-hindi-10000

该数据集包含了音频文件路径、文本、持续时间、语言、样本数、原文、标准化文本、说话者ID、场景、任务名称、性别、年龄段、职业类型、资格、地区、区域、州、职业、验证报告以及未过滤的原文和标准化文本等字段。数据集分为训练集等部分，提供了每个部分的大小和示例数量。

The dataset includes fields such as audio file path, text, duration, language, number of samples, verbatim, normalized text, speaker ID, scenario, task name, gender, age group, job type, qualification, area, district, state, occupation, verification report, and unsanitized verbatim and normalized text. The dataset is split into parts like training set, with each parts size and number of examples provided.