| CPC G10K 11/17885 (2018.01) [G06N 20/00 (2019.01); H04L 12/1822 (2013.01)] | 20 Claims |
|
1. A method comprising:
obtaining audio data from an endpoint participating in an online conference session, the audio data derived from audio captured at the endpoint;
obtaining a user selection indicating that a first type of noise reduction is to be applied to the audio captured at the endpoint;
classifying sounds in the audio data to identify at least one type of sound in response to obtaining the user selection that the first type of noise reduction is to be applied;
determining whether the at least one type of sound is on a list of allowed types of sounds, the list of allowed types of sounds indicating sounds that are to be transmitted to participants of the online conference session;
attenuating the at least one type of sound from the audio data when the at least one type of sound is not on the list of allowed types of sounds;
detecting a change in a sound environment associated with the endpoint; and
automatically activating a second type of noise reduction in response to detecting the change in the sound environment.
|