Voice command processing in low power devices

Patent metadata

Field	Value
Publication number	US-10601599-B2
Application number	US-201715859242-A
Country	US
Kind code	B2
Filing date	Dec 29, 2017
Priority date	Dec 29, 2017
Publication date	Mar 24, 2020
Grant date	Mar 24, 2020

Abstract

An audio processing device comprises audio input circuitry operable to receive audio input signals and to process the audio input signals to generate audio samples at a first rate. The audio processing device further comprises a first trigger engine operable to detect a keyword within the audio samples. Also, the audio processing device comprises a delay buffer operable to continuously receive and store the audio samples. The delay buffer is further operable to transfer the audio samples that are stored within the delay buffer to a host across a data bus at a second rate, which is faster than the first rate. Further, the delay buffer is operable to transfer the audio samples that are stored within the delay buffer to the host at the first rate, after the stored audio samples are transmitted.

First claim

What is claimed is:

1. A method for processing voice commands in a low power device comprising:
processing an audio input signal to generate a stream of voice enhanced audio samples at a first rate;
storing the voice enhanced audio samples in a delay buffer on a first-in, first-out basis;
detecting, using a first trigger engine, a keyword in the voice enhanced audio samples;
transmitting a wake up command to a host device to waken the host device from a sleep mode to an active move;
receiving a data transfer request from a host device;
transmitting the stored voice enhanced audio samples to an input buffer of the host device across a data bus at a second rate, wherein the second rate is greater than the first rate until a next voice enhanced audio sample is not stored in the delay buffer;
transmitting the generated stream of voice enhanced audio samples to the input buffer of the host device at the first rate until a stop command is received from the host device;
processing, by a voice command processor, the voice enhanced audio samples following the keyword to determine a command; and
executing the determined command.

2. The method of claim 1 , further comprising receiving the audio input signal from an audio sensor array, wherein the audio input signal is an audio signal.

3. The method of claim 1 wherein the delay buffer is configured to store voice enhanced audio samples representing T seconds of audio input data, wherein the T is a time period from receiving the audio input signal to the host device receiving first transmitted voice enhanced audio samples.

4. The method of claim 1 , further comprising sending a wake up command to the host device after detecting the keyword in the voice enhanced audio samples.

5. The method of claim 4 , further comprising:
receiving, by the host device, the wake up command;
transitioning, by the host device, from a sleep mode to an operating mode; and
transmitting, from the host device, a request for audio samples.

6. The method of claim 5 , further comprising receiving on the data bus the request for audio samples from the host device.

7. The method of claim 1 , wherein the second rate is a maximum available transmission rate for the data bus.

8. The method of claim 1 , further comprising:
receiving, by the host device, the transmitted voice enhanced audio samples; and
storing, by the host device, the voice enhanced audio samples in an input buffer.

9. The method of claim 8 , further comprising:
validating, by a second trigger engine of the host device, a presence of the keyword in the voice enhanced audio samples stored in the input buffer.

10. The method of claim 1 , further comprising:
detecting a period of inactivity of the host device;
sending, by a host device, the stop command; and
transitioning, by the host device, from an operational mode to a sleep mode.

11. An audio processing system for processing voice commands in a low power device comprising:
audio input circuitry operable to receive an audio input signal and generate a stream of voice enhanced audio samples having a first data rate;
a first trigger engine operable to detect a keyword in the stream of voice enhanced audio samples and transmit a wake up command to a host device across a data bus, wherein the host device is wakened from a sleep mode to an active mode;
a delay buffer operable to store the stream of voice enhanced audio samples on a first-in, first-out basis and, after receiving a data transfer request from the host device, transmit the stored voice enhanced audio samples to an input buffer of the host device at a second data rate that is greater than the first data rate until a next voice enhanced audio sample is not stored in the delay buffer;
wherein the delay buffer, after the stored voice enhanced audio samples are transmitted from the delay buffer to the input buffer of the host device, is operable to transmit the stream of voice enhanced audio samples to the input buffer of the host device at the first data rate until a stop command is received from the host device; and
wherein the host device further comprises a voice command processor operable to identify a voice command following the keyword in the voice enhanced audio samples stored in the input buffer and execute the identified voice command.

12. The audio processing system of claim 11 , further comprising an audio sensor array operable to sense sound from an environment and generate the audio input signal.

13. The audio processing system of claim 11 wherein the delay buffer is a memory operable to store a subset of the stream of voice enhanced audio samples representing at least T seconds of the audio input signal, wherein T is a time period between receiving the stream of voice enhanced audio samples to the host device receiving the first stored voice enhanced audio samples.

14. The audio processing system of claim 11 , wherein the first trigger engine is operable to send a wake up command to the host device after detecting the keyword in the stream of voice enhanced audio samples.

15. The audio processing system of claim 14 , wherein the host device comprises:
a host processor operable to receive the wake up command from the data bus, execute a wake up sequence, transmit on the data bus a request for voice enhanced audio samples and receive the stored voice enhanced audio samples.

16. The audio processing system of claim 11 , wherein the host device comprises an input buffer operable to store the voice enhanced audio samples received from the data bus.

17. The audio processing system of claim 16 , wherein the host device further comprises a second trigger engine operable to detect a presence of the keyword in the voice enhanced audio samples stored in the input buffer.

18. The audio processing system of claim 11 , wherein the host device is further operable to detect a period of inactivity, transmit a stop command across the data bus to stop transmission of the voice enhanced audio samples from the delay buffer, and transition to a sleep mode.

Assignees

• Synaptics Inc

Inventors

• Patel Manish J
• Renna Stanton
• Duligunti Vamshi
• Wang Johnny
• Chen Huanqi

Classifications

• G10L15/08

  Speech classification or search · CPC title

• G06F3/167

  Audio in a user interface, e.g. using voice commands for navigating, audio feedback · CPC title

• G10L2015/088

  Word spotting · CPC title

• G10L2015/223

  Execution procedure of a spoken command · CPC title

• H04L12/12 Primary

  Arrangements for remote connection or disconnection of substations or of equipment thereof · CPC title