CPC H04M 3/568 (2013.01) [H04M 2201/41 (2013.01)] | 18 Claims |
1. A method comprising:
obtaining, by a device, audio data from one or more past conferences;
generating, by the device, a prediction model that predicts when participants of a conference will speak based on the audio data from the one or more past conferences;
predicting, by the device and using the prediction model, a speech distribution for participants of a particular conference, wherein the speech distribution indicates when and how long each participant will speak during the particular conference; and
proactively optimizing, by the device, audio parameters of the particular conference based on its predicted speech distribution to prioritize audio output by a particular participant during the particular conference, wherein the particular participant is predicted to be a primary speaker during the particular conference.
|