This dataset was recorded in a vehicular noise environment, with 304 speakers participating, including 152 males and 152 females. All speakers involved in the recording were professionally selected to ensure standardized pronunciation and clear enunciation. The recorded texts cover information such as numbers, dates, times, and personal names.