The video app can clearly make a distinction between background noise and when a person speaks. Moreover, in video communication, the client can have some AI capabilities that can detect using image processing if someone's lips are moving and indicating they are speaking. Also, gesture sensing is possible like raising hands and indicating interest in speaking. With this in mind having a toggle option between manual mute and auto-mute could be a viable feature. This will enable hands-free conversations. If the auto mute is on then the system will keep on listening to sounds but won't transmit into the call. It will maintain a queue of a few milliseconds of voice data. Once the speaker tries to say something, the system will recognize and pull the voice waveforms from the queue, prepend it to the current waveforms, and transmit. A few milliseconds worth of wait might be needed to make it seamless. However, the video may stutter for those few milliseconds but it won't make much of an impact as it is minuscule and it is no stranger to VOIP video communications. The person will be automatically muted once the speaker stops and doesn't speak for say 5 seconds. And the process repeats each time a speaker tries to speak. None of the competitors have ever implemented this and I think it will be a great feature that will get popular in a few decades.