AudioLLMs/Multitask-National-Speech-Corpus-v1-extend

这是一个用于语音识别和对话系统的数据集，包含多个配置，每个配置都有音频上下文、指令、答案和其他属性。其他属性中包含了对话者的详细信息，如年龄、教育水平、民族、性别、职业等。数据集分为训练集和测试集，适用于构建和评估语音识别模型和对话系统。

This is a dataset for speech recognition and dialogue systems, containing multiple configurations. Each configuration includes audio context, instruction, answer, and other attributes. The other attributes contain detailed information about the speakers, such as age, education level, ethnicity, gender, occupation, etc. The dataset is divided into training and test sets, suitable for building and evaluating speech recognition models and dialogue systems.