Abstract: A microphone that includes a calibration signal generator that generates a calibration signal based upon a received invocation input. The calibration signal has known audio properties. The microphone also includes a transducer that converts sound energy into an electrical signal. An integrated circuit receives the electrical signal from the transducer. An external microphone interface provides the calibration signal to a device external to the microphone.

Abstract: Provided are systems and methods for utilizing digital microphones in low power keyword detection and noise suppression. An example method includes receiving a first acoustic signal representing at least one sound captured by a digital microphone. The first acoustic signal includes buffered data transmitted with a first clock frequency. The digital microphone may provide voice activity detection. The example method also includes receiving at least one second acoustic signal representing the at least one sound captured by a second microphone, the at least one second acoustic signal including real-time data. The first and second acoustic signals are provided to an audio processing system which may include noise suppression and keyword detection. The buffered portion may be sent with a higher, second clock frequency to eliminate a delay of the first acoustic signal from the second acoustic signal. Providing the signals may also include delaying the second acoustic signal.

Abstract: Systems and methods for acoustic echo cancellation are provided. An example method includes receiving a reference signal. The reference signal represents at least one sound captured inside an enclosure of a loudspeaker. The loudspeaker is operable to play back a far end signal. The method also includes receiving an acoustic signal. The acoustic signal represents at least one sound captured outside the enclosure of the loudspeaker. The acoustic signal includes at least a near end signal and the far end signal. The method enables attenuation, using the reference signal, the far end signal in the acoustic signal. The reference signal can be captured by a low sensitivity microphone placed inside the enclosure of the loudspeaker. The attenuation of the far end signal may include subtractive cancellation.

Abstract: An application specific integrated circuit (ASIC) is used with an acoustic device. An input clock signal is received. The frequency of the input clock signal is determined, and the frequency is indicative of one of a plurality of operational modes of the ASIC. Based upon the determined frequency, an amount current provided to one or more operational blocks of the ASIC is changed.

Abstract: A microelectromechanical system (MEMS) includes a diaphragm with a first surface and a second surface. The first surface is exposed to an environmental pressure. The second surface comprises a plurality of fingers extending from the second surface. The MEMS also includes a backplate comprising a plurality of voids. Each of the plurality of fingers extends into a respective one of the plurality of voids. The MEMS further includes an insulator between a portion of the diaphragm and a portion of the backplate. The diaphragm is configured to move with respect to the backplate in response to changes in the environmental pressure.

Abstract: Analog signals are received from a sound transducer. The analog signals are converted into digitized data. A determination is made as to whether voice activity exists within the digitized signal. Upon the detection of voice activity, an indication of voice activity is sent to a processing device. The indication is sent across a standard interface, and the standard interface is configured to be compatible to be coupled with a plurality of devices from potentially different manufacturers.

Abstract: Systems and methods for reducing false alarms in keyword detection are provided. An example method includes detecting a keyword in an acoustic signal. The acoustic signal can represent at least one captured sound. The method also includes acquiring an estimate of speech activity for a portion of the acoustic signal preceding the keyword. In some embodiments, the estimate includes an average of a voice activity detection output over frames of the acoustic signal within the portion preceding the keyword. If the estimate is less than a threshold, the method can accept the keyword detection. If the estimate is larger than the threshold, the method proceeds to reject the keyword detection.

Abstract: Systems and methods for audio monitoring and adaptation are provided. An example method includes monitoring an acoustic signal, representing at least one captured sound, inside at least one ear canal. The captured sound includes an audio for play back inside the ear canal. The acoustic signal can be analyzed to determine perceptual parameters including level of the acoustic signal, duration of the acoustic signal, inter-aural time difference (ITD), inter-aural level difference (ILD), seal quality, and environmental noise estimate. Based on the perceptual parameters, the played back audio is adapted to improve quality thereof. The adaptation includes regulating the volume of the acoustic signal, performing noise-dependent gain control on the acoustic signal, performing inter-aural temporal alignment and spectral equalization, and equalizing an acoustic response inside the ear canal.

Abstract: The disclosure relates to a microphone assembly including a multibit analog-to-digital converter configured to generate N-bit samples representative of a microphone signal. The microphone assembly also includes a first digital-to-digital converter configured to generate a corresponding M-bit digital signal based on N-bit digital samples, wherein N and M are positive integers and N>M. The microphone assembly may include a data interface configured to repeatedly receive samples of the M-bit digital signal and write bits of the M-bit digital signal to a data frame.

Abstract: An integrated armature and yoke for an acoustic receiver is disclosed. The armature includes a generally u-shaped member having a first portion arranged substantially parallel to a second deflecting portion. The yoke includes a central portion connected to the first portion, and first and second wall portions extending from opposite sides of the central portion toward the second deflecting portion. In a receiver, the second portion of the integrated armature and yoke is disposed through a receiver coil and extends into a space between permanent magnets, wherein an excitation signal applied to the coil causes the second deflecting portion of the armature to deflect between the magnets and to drive an acoustic sound generating diaphragm.

Abstract: Systems and methods for locating the end of a keyword in voice sensing are provided. An example method includes receiving an acoustic signal that includes a keyword portion immediately followed by a query portion. The acoustic signal represents at least one captured sound. The method further includes determining the end of the keyword portion. The method further includes, separating, using the end of the keyword portion, the query portion from the keyword portion of the acoustic signal. The method further includes providing the query portion, absent any part of the keyword portion, to an automatic speech recognition (ASR) system.

Abstract: A method for restoring distorted speech components of an audio signal distorted by a noise reduction or a noise cancellation includes determining distorted frequency regions and undistorted frequency regions in the audio signal. The distorted frequency regions include regions of the audio signal in which a speech distortion is present. Iterations are performed using a model to refine predictions of the audio signal at distorted frequency regions. The model is configured to modify the audio signal and may include deep neural network trained using spectral envelopes of clean or undamaged audio signals. Before each iteration, the audio signal at the undistorted frequency regions is restored to values of the audio signal prior to the first iteration; while the audio signal at distorted frequency regions is refined starting from zero at the first iteration. Iterations are ended when discrepancies of audio signal at undistorted frequency regions meet pre-defined criteria.

Abstract: A surface mount package for a micro-electro-mechanical system (MEMS) microphone die is disclosed. The surface mount package features a substrate with metal pads for surface mounting the package to a device's printed circuit board and for making electrical connections between the microphone package and the device's circuit board. The surface mount microphone package has a cover, and the MEMS microphone die is substrate-mounted and acoustically coupled to an acoustic port provided in the surface mount package. The substrate and the cover are joined together to form the MEMS microphone, and the substrate and cover cooperate to form an acoustic chamber for the substrate-mounted MEMS microphone die.

Abstract: Provided are systems and methods for microphone signal fusion. An example method commences with receiving a first and second signal representing sounds captured, respectively, by external and internal microphones. The internal microphone is located inside an ear canal and sealed for isolation from outside acoustic signals. The external microphone is located outside the ear canal. The first signal comprises a voice component. The second signal comprises a voice component modified by at least human tissue. The first and second signals are processed to obtain noise estimates. The voice component of the second signal is aligned with the voice component of the first signal. The first signal and the aligned voice component of the second signal are blended, based on the noise estimates, to generate an enhanced voice signal. Prior to aligning, the voice component of the second signal may be processed to emphasize high frequency content, improving effective alignment bandwidth.

Abstract: Provided are methods and systems for passive training for automatic speech recognition. An example method includes utilizing a first, speaker-independent model to detect a spoken keyword or a key phrase in spoken utterances. While utilizing the first model, a second model is passively trained to detect the spoken keyword or the key phrase in the spoken utterances using at least partially the spoken utterances. The second, speaker dependent model may utilize deep neural network (DNN) or convolutional neural network (CNN) techniques. In response to completion of the training, a switch is made from utilizing the first model to utilizing the second model to detect the spoken keyword or the key phrase in spoken utterances. While utilizing the second model, parameters associated therewith are updated using the spoken utterances in response to detecting the keyword or the key phrase in the spoken utterances. User authentication functionality may be provided.

Abstract: Systems and methods for multiplying a sparse matrix by a vector using a single instruction multiple data (SIMD) architecture are provided. An example method includes sorting rows of the sparse matrix by a number of non-zero elements in the rows to generate sorted rows. The sorted rows are split to generate groups of the sorted rows. The number of rows in each group of the sorted rows is equal to the number of rows updated in parallel. The method allows for packing the sorted rows in each of the groups to generate packed rows. Each of the packed rows within the same group has the same length. Per clock cycle, C elements of the packed rows and data for selecting elements of the vector are provided to computational units in the SIMD architecture, where C is the number of computational units.

Abstract: A microphone includes a base and a microelectromechanical system (MEMS) die and an integrated circuit (IC) disposed on the base. The microphone also includes a cover mounted on the base and covering the MEMS die and the IC. The cover includes an indented region or an inwardly drawn region that define a top port through which acoustic energy can enter the microphone and be incident on the MEMS die. The microphone also includes a filtering material disposed on the top port on an outside surface of the cover and within the indented region or the inwardly drawn region. The filtering material provides resistance to ingression of solid particles or liquids into the microphone.

Abstract: Systems and methods for stereo separation and directional suppression are provided. An example method includes receiving a first audio signal, representing sound captured by a first microphone associated with a first location, and a second audio signal, representing sound captured by a second microphone associated with a second location. The microphones comprise omni-directional microphones. The distance between the first and second microphones is limited by the size of a mobile device. A first channel signal of a stereo signal is generated by forming, based on the first and second audio signals, a first beam at the first location. A second channel signal of the stereo signal is generated by forming, based on the first and second audio signals, a second beam at the second location. First and second directions, associated respectively with the first and second beams, are fixed relative to a line between the first and second locations.

Abstract: A surface-mountable MEMS microphone comprising a MEMS microphone die and an application-specific integrated circuit (ASIC) mounted inside a surface-mountable package housing, and fully enclosed therein. The surface-mountable package is a single, self-contained housing that provides an electrical interface to external circuitry for the enclosed MEMS microphone die and the ASIC, and provides electrical, physical, and environmental protection for the MEMS microphone die and the ASIC. The surface-mountable package allows external acoustic energy to enter the package interior via one or more acoustic ports and impinge on the diaphragm of the MEMS microphone die. The cover of the surface-mountable package comprises an acoustic port with ingress protection to limit dust and particle intrusion. The ingress protection can be a formed member that is part of the cover of the surface-mountable package having various shapes, an internal shield, or a combination of both a formed member and internal shield.

Abstract: A device includes a substrate, strain gauges, and a controller coupled to the strain gauges. The substrate has a front surface and an opposing rear surface, the front surface including a button representation and the rear surface including a button area corresponding to the button representation. The strain gauges are mounted on the rear surface in proximity to the button area. The controller receives information indicating multiple electrical signal amplitudes, each of the electrical signal amplitudes corresponding to one strain gauge of the plurality of strain gauges, each electrical signal amplitude representing an amount of deformation of the corresponding strain gauge. The controller estimates a location of a pressure applied and/or the magnitude of force applied on the front surface of the substrate based on the received information. The controller also can receive location information of multiple user presses from a separate set of sensors.