This dataset was recorded in a car noise environment, with the involvement of 40 speakers, including 20 males and 20 females. All participants in the recording were professionally selected to ensure standard pronunciation and clear enunciation. The recorded text covers information on navigation, text messages, news, and other topics.