This dataset was recorded in a vehicular noise environment, with 303 speakers participating, including 148 males and 155 females. All speakers involved in the recording were carefully selected by professionals to ensure standardized pronunciation and clear articulation. The recorded texts cover information such as numbers, dates, times, and personal names.