This dataset was recorded in both quiet and noisy environments, with the participation of 200 speakers, including 107 males and 93 females. All speakers involved in the recording were professionally selected to ensure standardized pronunciation and clear enunciation. The recorded texts cover a range of information, including news updates, everyday conversations, and Twitter posts.