This dataset was recorded in quiet office/home environments, with the participation of 208 speakers, including 106 males and 102 females. All speakers involved in the recording were professionally selected to ensure standardized pronunciation and clear enunciation. The recorded texts cover everyday conversations.