This dataset was recorded in a quiet office environment, with 10 speakers participating, including 5 males and 5 females. All speakers involved in the recording were professionally selected to ensure standardized pronunciation and clear articulation. The recorded texts cover voicemail, paragraph dictation, and other information.