Abstract: The disclosure generally pertains to systems and methods for identifying a location of an occupant in a vehicle. In an example method, a processor deconvolves a vocal utterance by an occupant of a vehicle and also determines an angle of arrival of the vocal utterance. The location of the occupant in the vehicle is then identified by the processor based on evaluating the deconvolved vocal utterance and the angle of arrival of the vocal utterance. Deconvolving the vocal utterance can involve applying a cabin impulse response to the vocal utterance for eliminating undesirable effects that may be imposed upon the vocal utterance by acoustic characteristics of the cabin of the vehicle (echo, sound reflections, sound damping, reverberation etc.). In some applications, the processor may refer to a lookup table to estimate a location of the occupant in the vehicle.

Abstract: Systems and methods for assisting visually impaired passengers, e.g., blind passengers, using rideshare services is provided. The system may identify a visually impaired passenger from image data captured by one or more exterior cameras of a rideshare vehicle, e.g., by detection of use of a cane by the passenger. The system may then generate audible instructions that may be emitted via exterior and interior speakers of the vehicle or via a wirelessly paired headset worn by the passenger, to guide the passenger to the vehicle entrance and to an available passenger seat within the vehicle. The system further may detect obstacles or features in the path of the passenger, and emit audible alerts to guide the passenger around the obstacle or feature.

Abstract: A system for dynamically controlling an echo canceler includes a processor programmed to receive audio data from an audio sensor, to determine a present type of noise in the received audio data, based on the present type of noise, to determine a number of taps to be removed from an echo canceler, and to execute the echo canceler with a reduced number of taps based on the number of taps to be removed.

Abstract: The disclosure generally pertains to systems and methods for identifying a location of an occupant in a vehicle. In an example method, a processor deconvolves a vocal utterance by an occupant of a vehicle and also determines an angle of arrival of the vocal utterance. The location of the occupant in the vehicle is then identified by the processor based on evaluating the deconvolved vocal utterance and the angle of arrival of the vocal utterance. Deconvolving the vocal utterance can involve applying a cabin impulse response to the vocal utterance for eliminating undesirable effects that may be imposed upon the vocal utterance by acoustic characteristics of the cabin of the vehicle (echo, sound reflections, sound damping, reverberation etc.). In some applications, the processor may refer to a lookup table to estimate a location of the occupant in the vehicle.

Abstract: The disclosure describes systems and methods for processing audio signals in a vehicle to perform sound source separation. The sound source separation is performed using transfer functions and involves separation of the speech of multiple occupants. The separated speech can be used to isolate and correctly respond to a command to control vehicle systems.

Abstract: A system for dynamically controlling an echo canceler includes a processor programmed to receive audio data from an audio sensor, to determine a present type of noise in the received audio data, based on the present type of noise, to determine a number of taps to be removed from an echo canceler, and to execute the echo canceler with a reduced number of taps based on the number of taps to be removed.

Abstract: The disclosure describes systems and methods for processing audio signals in a vehicle to perform sound source separation. The sound source separation is performed using transfer functions and involves separation of the speech of multiple occupants. The separated speech can be used to isolate and correctly respond to a command to control vehicle systems.

Abstract: Detecting and isolating competing speech for voice controlled systems are provided herein. An example method includes isolating a voice command from a plurality of competing voice sound signals received internally or externally to a vehicle, wherein at least a portion of the plurality of competing voice sound signals are received using a sensor coupled with a window of the vehicle, and processing the voice command by the voice command system.

Abstract: Method and apparatus are disclosed for accelerometer-based external sound monitoring for voice controlled autonomous parking. An example vehicle includes a body control module, an infotainment head unit, and an autonomy unit. The example body control module communicatively couples to a mobile device. The example infotainment head unit communicatively couples to an accelerometer mounted on a window of the vehicle. The example autonomy unit autonomously parks the vehicle in response to a first key word detected by the accelerometer when (i) a button is activated on the infotainment head unit and (ii) a message from the mobile device is being periodically received.

Abstract: Method and apparatus are disclosed for monitoring of vehicle window vibrations for voice-command recognition. An example vehicle includes a window, an outer layer, a vibration sensor coupled to the window to detect audio vibrations, an audio actuator coupled to the outer layer to vibrate the outer layer, and a controller. The controller is to detect a voice command from a user via the vibration sensor, identify an audio response based upon the voice command, and emit the audio response to the user via the audio actuator.

Abstract: Method and apparatus are disclosed for speech recognition for vehicle voice commands. An example vehicle includes a microphone to collect a signal including a voice command, memory, and a controller. The controller is configured to determine an initial identification by feeding the signal into a first automatic speech recognition (ASR) engine and determine habits by feeding user history into a habits engine. The controller also is configured to identify the voice command by feeding the signal, the initial identification, and the habits into a second ASR engine. The controller also is configured to perform a vehicle function based on the voice command.

Abstract: Detecting and isolating competing speech for voice controlled systems are provided herein. An example method includes isolating a voice command from a plurality of competing voice sound signals received internally or externally to a vehicle, wherein at least a portion of the plurality of competing voice sound signals are received using a sensor coupled with a window of the vehicle, and processing the voice command by the voice command system.

Abstract: Method and apparatus are disclosed for accelerometer-based external sound monitoring for position aware autonomous parking. An example vehicle includes accelerometers mounted on windows of the vehicle, a computing platform communicatively coupled to the accelerometers, and an autonomy unit. The computing platform receives, via the accelerometers, signals indicative of a stop command associated with a voice. Additionally, the computing platform validates the stop command when the voice either is authorized or originates from within an emergency zone. The autonomy unit prevents autonomous motion of the vehicle in response to receiving the validated stop command.

Abstract: Method and apparatus are disclosed for accelerometer-based external sound monitoring for voice controlled autonomous parking. An example vehicle includes a body control module, an infotainment head unit, and an autonomy unit. The example body control module communicatively couples to a mobile device. The example infotainment head unit communicatively couples to an accelerometer mounted on a window of the vehicle. The example autonomy unit autonomously parks the vehicle in response to a first key word detected by the accelerometer when (i) a button is activated on the infotainment head unit and (ii) a message from the mobile device is being periodically received.

Abstract: Method and apparatus are disclosed for accelerometer-based external sound monitoring for voice controlled autonomous parking. An example vehicle includes a body control module, an infotainment head unit, and an autonomy unit. The example body control module communicatively couples to a mobile device. The example infotainment head unit communicatively couples to an accelerometer mounted on a window of the vehicle. The example autonomy unit autonomously parks the vehicle in response to a first key word detected by the accelerometer when (i) a button is activated on the infotainment head unit and (ii) a message from the mobile device is being periodically received.

Abstract: Method and apparatus are disclosed for accelerometer-based external sound monitoring during low speed maneuvers. An example vehicle includes accelerometers affixed to windows of the vehicle, speakers located inside the vehicle, and an infotainment head unit. Each of the speakers is uniquely associated with one of the accelerometers. The infotainment head unit, when a speed of the vehicle satisfies a threshold, selects a zone corresponding to a direction of travel of the vehicle. Additionally, the infotainment head unit plays signals captured by the accelerometers associated with the selected zone on the corresponding speakers.

Abstract: Method and apparatus are disclosed for monitoring of vehicle window vibrations for voice-command recognition. An example vehicle includes a window, an outer layer, a vibration sensor coupled to the window to detect audio vibrations, an audio actuator coupled to the outer layer to vibrate the outer layer, and a controller. The controller is to detect a voice command from a user via the vibration sensor, identify an audio response based upon the voice command, and emit the audio response to the user via the audio actuator.

Abstract: Method and apparatus are disclosed for speech recognition for vehicle voice commands. An example vehicle includes a microphone to collect a signal including a voice command, memory, and a controller. The controller is configured to determine an initial identification by feeding the signal into a first automatic speech recognition (ASR) engine and determine habits by feeding user history into a habits engine. The controller also is configured to identify the voice command by feeding the signal, the initial identification, and the habits into a second ASR engine. The controller also is configured to perform a vehicle function based on the voice command.

Abstract: Method and apparatus are disclosed for monitoring of vehicle window vibrations for voice-command recognition. An example vehicle includes a window, an outer layer, a vibration sensor coupled to the window to detect audio vibrations, an audio actuator coupled to the outer layer to vibrate the outer layer, and a controller. The controller is to detect a voice command from a user via the vibration sensor, identify an audio response based upon the voice command, and emit the audio response to the user via the audio actuator.

Abstract: A processor of a vehicle speech recognition system recognizes speech via domain-specific language and acoustic models. The processor further, in response to the acoustic model having a confidence score for recognized speech falling within a predetermined range defined relative to a confidence score for the domain-specific language model, recognizes speech via the acoustic model only.