Exploiting visual information for enhancing audio signals via source separation and beamforming

A method, comprising:

• extracting, from media content and by utilizing instructions from a memory that are executed by a processor, first metadata, wherein the first metadata indicates information that identifies an interferer generating noise in an environment, information that identifies a location of the interferer, and information that identifies what a user is doing;
• cancelling, based on the first metadata, an audio profile loaded based on a location of the user, and a user profile comprising audio data associated with the user, the noise from the environment;
• and
• adjusting, based on the audio profile and the user profile, an audio signal generated by the user.

A non-transitory computer-readable device comprising instructions, which when executed by a processor, cause the processor to perform operations comprising:

• obtaining, from media content, first metadata, wherein the first metadata indicates information that identifies an interferer generating noise in an environment, information that identifies a location of the interferer, and information that identifies what a user is doing;
• cancelling, based on the first metadata, an audio profile loaded based on a location of the user, and a user profile comprising audio data associated with the user, the noise from the environment;
• and
• adjusting, based on the audio profile and the user profile, an audio signal generated by the user.