This dataset was recorded in quiet office/home environments, with the participation of 110 speakers, including 55 males and 55 females. All speakers involved in the recording were professionally selected to ensure standardized pronunciation and clear enunciation. The recorded texts cover news and everyday conversations.