This dataset was recorded in a quiet office environment, with the participation of 50 speakers, evenly divided into 25 males and 25 females. All speakers involved in the recording were professionally selected to ensure standardized pronunciation and clear articulation. The recorded texts cover information such as numbers, dates, times, and personal names.