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: In one example, a headset may obtain, from a first microphone that is configured at a first location relative to an audio source, a first audio signal having a first audio level. The headset may further obtain, from a second microphone that is configured at a second location relative to the audio source, a second audio signal having a second audio level. The second location may be a greater distance from the audio source than the first location. The headset may determine a target audio level based on the second audio level. The headset may adjust the first audio signal to the target audio level to produce an adjusted first audio signal, and output the adjusted first audio signal.

Abstract: In one example, a first headset establishes a connection with a second headset that is associated with a target participant. The first headset obtains, via the connection, a first audio signal corresponding to speech of the target participant. Based on the first audio signal, one or more parameters associated with a position or a movement of a head of a user of the first headset, one or more head-related transfer functions associated with a shape of the head of the user, and a layout of the environment of the first headset, the first headset modifies the first audio signal to produce a first modified audio signal that corresponds to the speech of the target participant that would be present at the head of the user in absence of the first headset and the noise generated in the environment.

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.

Abstract: Systems, processes, devices, apparatuses, algorithms and computer readable medium for suppressing spatial interference using a dual microphone array for receiving, from a first microphone and a second microphone that are separated by a predefined distance, and that are configured to receive source signals, respective first and second microphone signals based on received source signals. A phase difference between the first and the second microphone signals is calculated based on the predefined distance. An angular distance between directions of arrival of the source signals and a desired capture direction is calculated based on the phase difference. Directional-filter coefficients are calculated based on the angular distance. Undesired source signals are filtered from an output based on the directional-filter coefficients.

Abstract: Systems, processes, devices, apparatuses, algorithms and computer readable medium for suppressing spatial interference using a dual microphone array for receiving, from a first microphone and a second microphone that are separated by a predefined distance, and that are configured to receive source signals, respective first and second microphone signals based on received source signals. A phase difference between the first and the second microphone signals is calculated based on the predefined distance. An angular distance between directions of arrival of the source signals and a desired capture direction is calculated based on the phase difference. Directional-filter coefficients are calculated based on the angular distance. Undesired source signals are filtered from an output based on the directional-filter coefficients.

Abstract: A telepresence video conference endpoint device includes spaced-apart microphone arrays each configured to transduce sound into corresponding sound signals. A processor receives the sound signals from the arrays and determines a direction-of-arrival (DOA) of sound at each array based on the set of sound signals from that array, determines if each array is blocked or unblocked based on the DOA determined for that array, selects an array among the arrays based on whether each array is determined to be blocked or unblocked, and perform subsequent sound processing based on one or more of the sound signals from the selected array.

Abstract: Systems, processes, devices, apparatuses, algorithms and computer readable medium for suppressing spatial interference using a dual microphone array for receiving, from a first microphone and a second microphone that are separated by a predefined distance, and that are configured to receive source signals, respective first and second microphone signals based on received source signals. A phase difference between the first and the second microphone signals is calculated based on the predefined distance. An angular distance between directions of arrival of the source signals and a desired capture direction is calculated based on the phase difference. Directional-filter coefficients are calculated based on the angular distance. Undesired source signals are filtered from an output based on the directional-filter coefficients.

Abstract: A telepresence video conference endpoint device includes spaced-apart microphone arrays each configured to transduce sound into corresponding sound signals. A processor receives the sound signals from the arrays and determines a direction-of-arrival (DOA) of sound at each array based on the set of sound signals from that array, determines if each array is blocked or unblocked based on the DOA determined for that array, selects an array among the arrays based on whether each array is determined to be blocked or unblocked, and perform subsequent sound processing based on one or more of the sound signals from the selected array.

Abstract: Systems, processes, devices, apparatuses, algorithms and computer readable medium for suppressing spatial interference using a dual microphone array for receiving, from a first microphone and a second microphone that are separated by a predefined distance, and that are configured to receive source signals, respective first and second microphone signals based on received source signals. A phase difference between the first and the second microphone signals is calculated based on the predefined distance. An angular distance between directions of arrival of the source signals and a desired capture direction is calculated based on the phase difference. Directional-filter coefficients are calculated based on the angular distance. Undesired source signals are filtered from an output based on the directional-filter coefficients.