Abstract: Systems and methods are provided for processing a voice input stream with interruptions and/or supplemental comments. Generally, a virtual voice assistant may receive an input stream with a first input comprising a voice query from a first voice and a second input comprising a secondary query from a second voice (e.g., an interruption or a supplement). The virtual assistant may determine that the second voice does not match the first voice, and then process the voice query to produce first results. Some embodiments may determine whether the secondary query is a supplement or an interruption and, e.g., choose to ignore an interruption or set aside a supplement if it may be used to help the search query. In some embodiments, results for the first query may be compared with results for the first query with a portion of the supplement.

Abstract: One embodiment of the present invention sets forth a technique for providing audio enhancement to a user of a listening device. The technique includes reproducing a first audio stream, such as an audio stream associated with a media player. The technique further includes detecting a voice trigger. The voice trigger may be associated with a name of a user of the listening device. The technique further includes pausing or attenuating the first audio stream and reproducing a second audio stream associated with ambient sound in response to detecting the voice trigger.

Abstract: Approaches are disclosed for generating auditory scenes. A computing device includes a wireless network interface and a processor. The processor is configured to receive, via a microphone, a first auditory signal that includes a first plurality of voice components. The processor is further configured to receive a request to at least partially suppress a first voice component included in the first plurality of voice components. The processor is further configured to generate a second auditory signal that includes the first plurality of voice components with the first voice component at least partially suppressed. The processor is further configured to transmit the second auditory signal to a speaker for output.

Abstract: A personal safety system detects imminent danger or other event of interest and then modifies a sound panorama to focus a user's attention towards the direction of the danger or other event of interest. The personal safety system may isolate sounds originating from the direction of the danger or other event of interest, collapse the sound panorama towards the direction of the danger or other event of interest, or compress the sound panorama to align with the direction of the danger or other event of interest. The personal safety system may be integrated into an automobile or a wearable system physically attached to the user. Advantageously, the attention of the user may be drawn towards imminent danger or other events of interest without introducing additional sensory information to the user, thereby reducing the likelihood of startling or distracting the user.

Abstract: A universal device controller is configured to operate as a proxy mechanism for interacting with a plurality of different devices wirelessly coupled to the universal device controller. When a user of the universal device controller verbally expresses a command from a universal command language understood by the device, the universal device controller analyzes the verbal expression of the user and identifies a wireless device capable of executing the associated command. The universal device controller then identifies an API for interacting with the identified device and translates the verbal expression into a device command that the identified device is capable of executing. The universal device controller then issues the device command to the identified device.

Abstract: A recipient computing device can receive a speech utterance to be processed by speech recognition and segment the speech utterance into two or more speech utterance segments, each of which can be to one of a plurality of available speech recognizers. A first one of the plurality of available speech recognizers can be implemented on a separate computing device accessible via a data network. A first segment can be processed by the first recognizer and the results of the processing returned to the recipient computing device, and a second segment can be processed by a second recognizer implemented at the recipient computing device.

Abstract: In one embodiment, an apparatus for adaptively interacting with a driver via voice interaction is provided. The apparatus includes a computational models block and an adaptive interactive voice system. The computational models block is configured to receive driver related parameters, vehicle related parameters, and vehicle environment parameters from a plurality of sensors. The computational models block is further configured to generate a driver state model based on the driver related parameters and to generate a vehicle state model based on the vehicle related parameter. The computational models block is further configured to generate a vehicle environment state model based on the vehicle environment parameters. The adaptive interactive voice system is configured to generate a voice output based on a driver's situation and context as indicated on information included within at least one of the driver state model, the vehicle state model, and the vehicle environment state model.

Abstract: An audio processing system may selectively identify certain environmental sounds and playing back these sounds, or a representation of these sounds, in the vehicle's cabin. The audio processing system may filter the environmental sounds to identify a particular sound that matches an event such as a bouncing ball, squealing tires, footsteps, and the like. The audio processing system may then provide an audio alert to an occupant in the vehicle. For example, the system may process the identified sound (e.g., amplify and/or isolate the sound) and use a speaker to output the processed sound into the interior of the vehicle. In another embodiment, the audio processing system may use environmental sounds as an audio masking sound for creating privacy zones within the vehicle. The audio processing system may filter the environmental sounds to identify a continuous sound which is then output to generate the privacy zones.

Abstract: A product for detecting people approaching an isolated user, discouraging the detected people from disturbing the user, and for taking messages from the approaching people. The product may be embodied in noise-canceling headphones that the user is wearing, for example. The product can include a camera and a microphone that can detect approaching people. The product can also include an externally-facing speaker that can play an audio message to the approaching person to attempt to discourage the person from disturbing the user. The product can also prompt the person to leave a message for the user and then record the subsequent message. The user can review the message at a later time.

Abstract: A personal safety system detects imminent danger or other event of interest and then modifies a sound panorama to focus a user's attention towards the direction of the danger or other event of interest. The personal safety system may isolate sounds originating from the direction of the danger or other event of interest, collapse the sound panorama towards the direction of the danger or other event of interest, or compress the sound panorama to align with the direction of the danger or other event of interest. The personal safety system may be integrated into an automobile or a wearable system physically attached to the user. Advantageously, the attention of the user may be drawn towards imminent danger or other events of interest without introducing additional sensory information to the user, thereby reducing the likelihood of startling or distracting the user.

Abstract: A recipient computing device can receive a speech utterance to be processed by speech recognition and segment the speech utterance into two or more speech utterance segments, each of which can be to one of a plurality of available speech recognizers. A first one of the plurality of available speech recognizers can be implemented on a separate computing device accessible via a data network. A first segment can be processed by the first recognizer and the results of the processing returned to the recipient computing device, and a second segment can be processed by a second recognizer implemented at the recipient computing device.

Abstract: A method for generating an auditory environment for a user may include receiving a signal representing an ambient auditory environment of the user, processing the signal using a microprocessor to identify at least one of a plurality of types of sounds in the ambient auditory environment, receiving user preferences corresponding to each of the plurality of types of sounds, modifying the signal for each type of sound in the ambient auditory environment based on the corresponding user preference, and outputting the modified signal to at least one speaker to generate the auditory environment for the user. A system may include a wearable device having speakers, microphones, and various other sensors to detect a noise context. A microprocessor processes ambient sounds and generates modified audio signals using attenuation, amplification, cancellation, and/or equalization based on user preferences associated with particular types of sounds.

Abstract: A recipient computing device can receive a speech utterance to be processed by speech recognition and segment the speech utterance into two or more speech utterance segments, each of which can be to one of a plurality of available speech recognizers. A first one of the plurality of available speech recognizers can be implemented on a separate computing device accessible via a data network. A first segment can be processed by the first recognizer and the results of the processing returned to the recipient computing device, and a second segment can be processed by a second recognizer implemented at the recipient computing device.

Abstract: A personal safety system detects imminent danger or other event of interest and then modifies a sound panorama to focus a user's attention towards the direction of the danger or other event of interest. The personal safety system may isolate sounds originating from the direction of the danger or other event of interest, collapse the sound panorama towards the direction of the danger or other event of interest, or compress the sound panorama to align with the direction of the danger or other event of interest. The personal safety system may be integrated into an automobile or a wearable system physically attached to the user. Advantageously, the attention of the user may be drawn towards imminent danger or other events of interest without introducing additional sensory information to the user, thereby reducing the likelihood of startling or distracting the user.

Abstract: A system, and method for communicating navigation information to a vehicle operator through actuators in a steering mechanism for a vehicle. The steering mechanism can include multiple actuators along or in a surface a vehicle operator touches to steer the vehicle. Different numbers of actuators can be actuated under an operator's hand to communicate the severity of an upcoming turn. Alternatively, an actuator can be actuated by different amounts to communicate the severity of an upcoming turn. Proximity of an upcoming turn can be communicated by cycling actuators at different rates for different turns. By actuating actuators under an operator's left hand or the left portion of a hand, a left-hand turn can be communicated. Conversely, by actuating actuators under an operator's right hand or the right portion of a hand, a right-hand turn can be communicated.

Abstract: An audio processing system may selectively identify certain environmental sounds and playing back these sounds, or a representation of these sounds, in the vehicle's cabin. The audio processing system may filter the environmental sounds to identify a particular sound that matches an event such as a bouncing ball, squealing tires, footsteps, and the like. The audio processing system may then provide an audio alert to an occupant in the vehicle. For example, the system may process the identified sound (e.g., amplify and/or isolate the sound) and use a speaker to output the processed sound into the interior of the vehicle. In another embodiment, the audio processing system may use environmental sounds as an audio masking sound for creating privacy zones within the vehicle. The audio processing system may filter the environmental sounds to identify a continuous sound which is then output to generate the privacy zones.

Abstract: A roadway projection system integrated into a vehicle is configured to identify a specific driving scenario encountered by a driver of the vehicle, and to then project an image onto a roadway along which the vehicle travels based on that scenario. The image is intended to provide guidance to the driver in negotiating the identified scenario and/or to visualize potential driving actions of the driver for the benefit of others. The image could be, for example, and without limitation, an indicator that the driver should follow to perform a specific driving or navigation action, or a preference that the driver wishes to share with other drivers. In addition, the roadway projection system may detect images projected by other roadway projection systems in other vehicles and to determine mutual preferences shared between drivers. When a shared preference is violated, the roadway projection system may alert the driver of the vehicle.

Abstract: An audio processing system may selectively identify certain environmental sounds and playing back these sounds, or a representation of these sounds, in the vehicle's cabin. The audio processing system may filter the environmental sounds to identify a particular sound that matches an event such as a bouncing ball, squealing tires, footsteps, and the like. The audio processing system may then provide an audio alert to an occupant in the vehicle. For example, the system may process the identified sound (e.g., amplify and/or isolate the sound) and use a speaker to output the processed sound into the interior of the vehicle. In another embodiment, the audio processing system may use environmental sounds as an audio masking sound for creating privacy zones within the vehicle. The audio processing system may filter the environmental sounds to identify a continuous sound which is then output to generate the privacy zones.

Abstract: In one embodiment, an apparatus for adaptively interacting with a driver via voice interaction is provided. The apparatus includes a computational models block and an adaptive interactive voice system. The computational models block is configured to receive driver related parameters, vehicle related parameters, and vehicle environment parameters from a plurality of sensors. The computational models block is further configured to generate a driver state model based on the driver related parameters and to generate a vehicle state model based on the vehicle related parameter. The computational models block is further configured to generate a vehicle environment state model based on the vehicle environment parameters. The adaptive interactive voice system is configured to generate a voice output based on a driver's situation and context as indicated on information included within at least one of the driver state model, the vehicle state model, and the vehicle environment state model.

Abstract: Approaches are disclosed for generating auditory scenes. A computing device includes a wireless network interface and a processor. The processor is configured to receive, via a microphone, a first auditory signal that includes a first plurality of voice components. The processor is further configured to receive a request to at least partially suppress a first voice component included in the first plurality of voice components. The processor is further configured to generate a second auditory signal that includes the first plurality of voice components with the first voice component at least partially suppressed. The processor is further configured to transmit the second auditory signal to a speaker for output.