This dataset was recorded in a vehicular noise environment, with the participation of 316 speakers, including 156 males and 160 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.