A system and method for improving the performance of a hands-free voice user interface system while minimizing the computational complexity without sacrificing performance. Specifically, when estimating the location of the talker for the purpose of steering a directional beam in the direction of the active talker. A hands-free voice user interface system requires a clean signal to be streamed to the cloud for recognition. One way to improve the speech signal is to estimate where the talker is and steer a beam in the direction of the active talker. To locate the talker to a localized position, a direction of arrival estimator (DOA) algorithm is used. DoA generally requires noise and echo free signal for optimal estimation, but it is computationally expensive to run audio pre-processing such as an acoustic echo cancellation for each microphone in microphone array. To reduce computational complexity, the system and method extract certain range of frequency and operate pre-processing only on the selected frequency. By properly selecting the frequency range, it does not degrade DoA accuracy while significantly reducing computational complexity.

The systems, devices, and processes described herein may identify a beam of a voice-controlled device that is directed toward a reflective surface, such as a wall. The beams may be created by a beamformer. An acoustic echo canceller (AEC) may create filter coefficients for a reference sound. The filter coefficients may be analyzed to identify beams that include multiple peaks. The multiple peaks may indicate presence of one or more reflective surfaces. Using the amplitude and the time delay between the peaks, the device may determine that it is close to a reflective surface in a direction of the beam.

The systems, devices, and processes described herein may identify a beam of a voice-controlled device that is directed toward a reflective surface, such as a wall. The beams may be created by a beamformer. An acoustic echo canceller (AEC) may create filter coefficients for a reference sound. The filter coefficients may be analyzed to identify beams that include multiple peaks. The multiple peaks may indicate presence of one or more reflective surfaces. Using the amplitude and the time delay between the peaks, the device may determine that it is close to a reflective surface in a direction of the beam.

Systems and methods are disclosed for autonomous joint detection-classification of acoustic sources of interest. Localization and tracking from unmanned marine vehicles are also described. Based on receiving acoustic signals originating above or below the surface, a processor can process the acoustic signals to determine the target of interest associated with the acoustic signal. The methods and systems autonomously and jointly detect and classify a target of interest. A target track can be generated corresponding to the locations of the detected target of interest. A classifier can be used representing spectral characteristics of a target of interest.

Techniques are provided for audio-based detection and tracking of an acoustic source. A methodology implementing the techniques according to an embodiment includes generating acoustic signal spectra from signals provided by a microphone array, and performing beamforming on the acoustic signal spectra to generate beam signal spectra, using time-frequency masks to reduce noise. The method also includes detecting, by a deep neural network (DNN) classifier, an acoustic event, associated with the acoustic source, in the beam signal spectra. The DNN is trained on acoustic features associated with the acoustic event. The method further includes performing pattern extraction, in response to the detection, to identify time-frequency bins of the acoustic signal spectra that are associated with the acoustic event, and estimating a motion direction of the source relative to the array of microphones based on Doppler frequency shift of the acoustic event calculated from the time-frequency bins of the extracted pattern.

In accordance with an embodiment a package includes: a package structure which defines inner surfaces delimiting an inner volume and outer surfaces directed towards an exterior of the package; at least one acoustic sensor element applied to at least one of the inner surfaces, to convert acoustic waves arriving from the exterior of the package into acoustic information in the form of electric signals; a plurality of millimeter wave sensing elements applied to at least one of the outer surfaces, to receive reflected radar signals from objects in the exterior of the package; and a circuitry applied to at least one of the inner surfaces of the package structure, wherein the circuitry is electrically connected to the at least one acoustic sensor element and the plurality of millimeter wave sensing elements to process the acoustic information and the reflected radar signals.

A voice communication system (100) is described. The voice communication system (100) may include an audio engine (112) and a mapping engine (114). The audio engine (112) may cancel ambient noise from a plurality of acoustic signals, to obtain a first set of signals. Further, the audio engine (112) may determine a number of acoustic signals in the first set of acoustic signals and a number of sound sources pertaining to the first set of acoustic signals. The mapping engine (114) may suppress noise from each of the first set of acoustic signals to obtain a noise free set of acoustic signals. In addition, the mapping engine (114) may identify a primary acoustic signal from amongst the noise free set of acoustic signals by mapping each noise free acoustic signal to a corresponding sound source.

A voice communication system (100) is described. The voice communication system (100) may include an audio engine (112) and a mapping engine (114). The audio engine (112) may cancel ambient noise from a plurality of acoustic signals, to obtain a first set of signals. Further, the audio engine (112) may determine a number of acoustic signals in the first set of acoustic signals and a number of sound sources pertaining to the first set of acoustic signals. The mapping engine (114) may suppress noise from each of the first set of acoustic signals to obtain a noise free set of acoustic signals. In addition, the mapping engine (114) may identify a primary acoustic signal from amongst the noise free set of acoustic signals by mapping each noise free acoustic signal to a corresponding sound source.

Systems and methods for detecting and acknowledging audio transmissions containing data. In one embodiment, a method is presented that includes receiving multiple audio signals that are detected by multiple receivers from within a service area. A first audio transmission may be detected in a first subset of the audio signal that are received by a first subset of the receivers. The first subset of the receivers may be positioned to receive audio transmissions from computing devices located within a first portion of the service area. At least one transmitter may be identified that is positioned to transmit audio transmissions to computing devices located within at least a subset of the first portion of the service area. A second audio transmission may be transmitted using the at least one first transmitter.

Technologies for a fabric acoustic sensor are disclosed. The fabric acoustic sensor includes a conductive thread and a non-conductive thread, which form a diaphragm that vibrates in response to a sound wave. As a result of the vibration, the conductive thread stretches, and a resistance of the conductive thread varies. The change in resistance is measured by a compute device, and the compute device may determine the sound wave based on the change in resistance. In some embodiments, the fabric acoustic sensor may be used to monitor a heart rate, locate an object, and/or provide an input for noise cancellation.