This dataset was recorded in quiet office/home environments, with the participation of 51 speakers, including 31 males and 20 females. All speakers involved in the recording were professionally selected to ensure standardized pronunciation and clear enunciation. The recorded texts cover customer service dialogues, "spot the difference" games, and everyday conversations.