Abstract: Embodiments describe a method for speech endpoint detection including receiving identification data for a first state associated with a first frame of speech data from a WFST language model, determining that the first frame of the speech data includes silence data, incrementing a silence counter associated with the first state, copying a value of the silence counter of the first state to a corresponding silence counter field in a second state associated with the first state in an active state list, and determining that the value of the silence counter for the first state is above a silence threshold. The method further includes, determining that an endpoint of the speech has occurred in response to determining that the silence counter is above the silence threshold, and outputting text data representing a plurality of words determined from the speech data that was received prior to the endpoint.

Abstract: Embodiments describe a method for decoding speech including receiving speech input at an audio input device, generating speech data that is a digital representation of the speech input; extracting acoustic features of the speech data, assigning acoustic scores to the acoustic features, receiving data representing the acoustic features and the acoustic scores, decoding the data representing the acoustic features into a word, having a word score, by referencing a WFST language model, modifying the word score into a new word score based on a personalized grammar model stored in the external memory device, the processor is separate from and external to the WFST accelerator, and determining an intent represented by a plurality of words outputted by the WFST accelerator, where the plurality of words include the word and the new word score.

Abstract: Embodiments describe a method for speech endpoint detection including receiving identification data for a first state associated with a first frame of speech data from a WFST language model, determining that the first frame of the speech data includes silence data, incrementing a silence counter associated with the first state, copying a value of the silence counter of the first state to a corresponding silence counter field in a second state associated with the first state in an active state list, and determining that the value of the silence counter for the first state is above a silence threshold. The method further includes, determining that an endpoint of the speech has occurred in response to determining that the silence counter is above the silence threshold, and outputting text data representing a plurality of words determined from the speech data that was received prior to the endpoint.

Abstract: This disclosure describes, in part, voice-controlled light dimmers that act as voice-controlled endpoints at which users may provide voice commands. These light dimmers include a front panel module coupled to a power module using a hardware interface. The front panel module may receive input from a user indicating commands for controlling appliances, and send communications to the power module using the hardware interface to control the appliances. In some examples, the communications involve encrypted data sent using an inter-integrated circuit (I2C) protocol using the hardware interface to an electrically isolated power module. The power provided to the appliances may be controlled by the power module of the voice-controlled light dimmer.

Abstract: A modular controller may be mounted in an opening, such as a standard single wide or double wide electrical junction box, in a wall or other surface. The modular controller may include a power module and a front module. The power module may be mounted in the opening of the surface, and may be configured to provide electrical power to the front module. The front module may be detachably coupleable to the power module. The front module may be configured to receive audio commands, gesture commands, and/or presence input corresponding to a desired action, and may cause the action to be performed by a device of the front module and/or an external device. The front module may include various devices (e.g., components) capable of providing various functionalities, and may be selected for coupling to a power module in a particular location based at least in part on the functionalities.

Abstract: An echo cancellation system that synchronizes output audio data with input audio data in a heterogeneous system. The system may append a most recent outgoing audio data frame to an incoming audio data frame to synchronize a receiving buffer with a transmission buffer. By synchronizing the receiving buffer with the transmission buffer, the incoming audio data frames may be associated with corresponding outgoing audio data frames. After synchronizing the incoming audio data frames and the outgoing audio data frames, the system may perform Acoustic Echo Cancellation by removing the outgoing audio data frames from the incoming audio data frames.

Abstract: This disclosure describes, in part, voice-controlled light dimmers that act as voice-controlled endpoints at which users may provide voice commands. These light dimmers include a front panel module coupled to a power module using a hardware interface. The front panel module may receive input from a user indicating commands for controlling appliances, and send communications to the power module using the hardware interface to control the appliances. In some examples, the communications involve encrypted data sent using an inter-integrated circuit (I2C) protocol using the hardware interface to an electrically isolated power module. The power provided to the appliances may be controlled by the power module of the voice-controlled light dimmer.

Abstract: This disclosure describes voice-enabled modular devices that act as functional, voice-controlled endpoints at which users may provide voice commands. The voice-enabled modular devices may include a power unit which may be installed into a wall permanently to serve as a power source, and a front-panel module which detachably couples to the power unit via a hardware interface. The hardware interface may allow the front-panel module and power unit to communicate with each other, and may further allow the power unit to provide power to the front-panel module. The power unit of the voice-enabled modular device described herein may be configured to provide power to multiple front-panel modules having different capabilities, and potentially different power requirements, thereby supporting multiple front-panel modules to increase the functionality of the voice-enabled modular devices.

Abstract: A modular controller may be mounted in an opening, such as a standard single wide or double wide electrical junction box, in a wall or other surface. The modular controller may include a power module and a front module. The power module may be mounted in the opening of the surface, and may be configured to provide electrical power to the front module. The front module may be detachably coupleable to the power module. The front module may be configured to receive audio commands, gesture commands, and/or presence input corresponding to a desired action, and may cause the action to be performed by a device of the front module and/or an external device. The front module may include various devices (e.g., components) capable of providing various functionalities, and may be selected for coupling to a power module in a particular location based at least in part on the functionalities.

Abstract: An echo cancellation system that synchronizes output audio data with input audio data in a heterogeneous system. The system may increment a counter as outgoing audio frames are sent to a digital-to-analog converter in a speaker. As incoming audio frames are received by an analog-to-digital converter in a microphone, the system may copy contents of the counter into the incoming audio frames. Based on the contents of the counter, the incoming audio frames may be associated with corresponding outgoing audio frames. After synchronizing the incoming audio frames and the outgoing audio frames, the system may perform Acoustic Echo Cancellation by removing the outgoing audio frames from the incoming audio frames.

Abstract: An ultrasonic finger print authentication system that generates a three-dimensional representation of a fingerprint. High frequency sound waves are used as a medium for imaging finer details of the patterns on a fingertip, including the ridge and valley formations. Multiple frequencies and beam-forming are used to quickly map the ridges. Acoustic resonance is used to determine the depths of the intervening valleys.