This dataset was recorded in quiet office and home environments, with 150 speakers participating, including 106 males and 44 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 sentences, alphanumeric combinations of personal names, and place names.