A method, computer system, and a computer program product for sound management is provided. The present invention may include generating a multidimensional risk vector relating to a current contextual activity of a user. The present invention may include dynamically monitoring at least one sound generating device within a predefined threshold distance of a chatbot of the user. The present invention may include determining that an audio adjustment of the at least one sound generating device is warranted. The present invention may include performing the audio adjustment.