Abstract: A method is provided that is performed at a system including multiple speech collectors to collect speech from a talker to produce corresponding ones of multiple audio signals that each convey speech energy: for each audio signal: separating high-frequency speech energy from low-frequency speech energy; and determining a first energy level of the high-frequency speech energy; and determining a preferred audio signal among the multiple audio signals for subsequent processing at least based on the first energy level of each audio signal.

Abstract: In various embodiments, a collaboration endpoint transmits an ultrasonic signal into an area based on a frequency sweep. The collaboration endpoint receives a reflected signal that comprises the ultrasonic signal reflected off an object located in the area. The collaboration endpoint detects, based on the reflected signal, the object as being a potential user. The collaboration endpoint determines, based on the reflected signal, a distance from the collaboration endpoint to the potential user. The collaboration endpoint initiates, based on the distance from the collaboration endpoint to the potential user, a wake-up sequence of the collaboration endpoint to exit a standby mode.

Abstract: In various embodiments, a collaboration endpoint transmits an ultrasonic signal into an area based on a frequency sweep. The collaboration endpoint receives a reflected signal that comprises the ultrasonic signal reflected off an object located in the area. The collaboration endpoint detects, based on the reflected signal, the object as being a potential user. The collaboration endpoint determines, based on the reflected signal, a distance from the collaboration endpoint to the potential user. The collaboration endpoint initiates, based on the distance from the collaboration endpoint to the potential user, a wake-up sequence of the collaboration endpoint to exit a standby mode.

Abstract: An endpoint deployed in a room may analyze a video output of a video camera disposed within the room for an image of a microphone disposed within the room. Based on the analyzing, the endpoint may then detect the image of the microphone within the video output of the video camera. The endpoint may then determine if the microphone is receiving speech from an active speaking participant, who may be one of the one or more participants that are disposed within the room. In the event the microphone is receiving speech from the active speaking participant, the endpoint may then display on a video display the video output of the video camera that has been altered to show a framing of the active speaking participant.

Abstract: In various embodiments, a collaboration endpoint transmits an ultrasonic signal into an area. The collaboration endpoint receives a reflected signal that comprises the ultrasonic signal reflected off an object located in the area. The collaboration endpoint detects, based on the reflected signal, the object as being a potential user. The collaboration endpoint determines, based on the reflected signal, a distance from the collaboration endpoint to the potential user. The collaboration endpoint initiates, based on the distance from the collaboration endpoint to the potential user, a wake-up sequence of the collaboration endpoint to exit a standby mode.

Abstract: This disclosure relates to solutions for eliminating undesired audio artifacts, such as background noises, on an audio channel. A process for implementing the technology can include receiving a set of audio segments, analyzing the segments using a first ML model to identify a first probability of unwanted background noises in the segments, and if the first probability exceeds a threshold, analyzing the segments using a second ML model to determine a second probability that the one or more background features exist in the segments. In some aspects, the process can include attenuating audio artifacts in the segments, if the second probability exceeds a second threshold. In some implementations, dynamic time stretching and shrinking can be applied to the noise attenuation. Systems and machine-readable media are also provided.

Abstract: An endpoint deployed in a room may analyze a video output of a video camera disposed within the room for an image of a microphone disposed within the room. Based on the analyzing, the endpoint may then detect the image of the microphone within the video output of the video camera. The endpoint may then determine if the microphone is receiving speech from an active speaking participant, who may be one of the one or more participants that are disposed within the room. In the event the microphone is receiving speech from the active speaking participant, the endpoint may then display on a video display the video output of the video camera that has been altered to show a framing of the active speaking participant.

Abstract: This disclosure relates to solutions for eliminating undesired audio artifacts, such as background noises, on an audio channel. A process for implementing the technology can include receiving a set of audio segments, analyzing the segments using a first ML model to identify a first probability of unwanted background noises in the segments, and if the first probability exceeds a threshold, analyzing the segments using a second ML model to determine a second probability that the one or more background features exist in the segments. In some aspects, the process can include attenuating audio artifacts in the segments, if the second probability exceeds a second threshold. In some implementations, dynamic time stretching and shrinking can be applied to the noise attenuation. Systems and machine-readable media are also provided.

Abstract: This disclosure relates to solutions for eliminating undesired audio artifacts, such as background noises, on an audio channel. A process for implementing the technology can include receiving a set of audio segments, analyzing the segments using a first ML model to identify a first probability of unwanted background noises in the segments, and if the first probability exceeds a threshold, analyzing the segments using a second ML model to determine a second probability that the one or more background features exist in the segments. In some aspects, the process can include attenuating audio artifacts in the segments, if the second probability exceeds a second threshold. In some implementations, dynamic time stretching and shrinking can be applied to the noise attenuation. Systems and machine-readable media are also provided.