Systems and methods for removing interference for audio pattern recognition

• US 10,468,020 B2
• Filed: 09/26/2017
• Issued: 11/05/2019
• Est. Priority Date: 06/06/2017
• Status: Active Grant

• Current Assignee: Cypress Semiconductor Corporation (Infineon Technologies AG)
• Original Assignee: Cypress Semiconductor Corporation (Infineon Technologies AG)
• Inventors: Pandey, Ashutosh, Zopf, Robert
• Primary Examiner(s): He, Jialong
• Application Number:
  US15/716,173
• Publication Number:
  US 20180350357A1
• Time in Patent Office: 770 Days
• CPC Class Codes: G10L 15/00 Speech recognition G10L17/00 takes precedence
• CPC Class Codes: G10L 15/20 Speech recognition techniques specially adapted for robustness in adverse environments, e.g. in noise, of stress induced speech G10L21/02 takes precedence
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue
• CPC Class Codes: G10L 15/30 Distributed recognition, e.g. in client-server systems, for mobile phones or network applications
• CPC Class Codes: G10L 19/005 Correction of errors induced by the transmission channel, if related to the coding algorithm
• CPC Class Codes: G10L 2015/223 Execution procedure of a spoken command
• CPC Class Codes: G10L 2021/02082 the noise being echo, reverberation of the speech
• CPC Class Codes: G10L 2021/02166 Microphone arrays; Beamforming
• CPC Class Codes: G10L 2025/783 based on threshold decision
• CPC Class Codes: G10L 21/0208 Noise filtering
• CPC Class Codes: G10L 21/0232 Processing in the frequency domain
• CPC Class Codes: G10L 25/84 for discriminating voice from noise
• CPC Class Codes: H04M 9/082 using echo cancellers echo cancellers per se H04B3/23

Abstract

An example apparatus provides an input signal based on sound waves received by one or more microphones. The input signal includes a voice command component and one or more interference components. The apparatus receives audio data over one or more computer networks and the audio data corresponds to the one or more interference components. The apparatus uses the audio data to remove a portion of the one or more interference components from the input signal to generate an output signal, and provides the output signal, as an estimate of the voice command component, for speech recognition.

Claims

1. A method comprising:

providing an input signal based on sound waves received by one or more microphones, the input signal including a voice command component and one or more interference components;

buffering the input signal;

detecting an indication to receive audio data;

receiving the audio data via one or more computer networks, the audio data corresponding to the one or more interference components, wherein the receiving of the audio data is responsive to the detecting of the indication, and wherein the buffering of the input signal includes buffering the input signal at least until the receiving of the audio data;

using the audio data, removing at least a portion of the one or more interference components from the input signal to generate an output signal; and

providing the output signal, as an estimate of the voice command component, for speech recognition.

2. The method of claim 1, wherein the using of the audio data includes generating a reference signal by combining first audio data of the audio data with second audio data of the audio data, wherein the first audio data corresponds to a first interference component of the one or more interference components, and the second audio data corresponds to a second interference component of the one or more interference components, wherein removing at least the portion of the one or more interference components from the input signal comprises subtracting the reference signal from the input signal.

3. The method of claim 2, wherein removing at least the portion of the one or more interference components from the input signal includes combining at least portion of the audio data with at least a portion of the input signal.

4. The method of claim 1, wherein removing at least the portion of the one or more interference components from the input signal includes comparing at least portion of the audio data with at least a portion of the input signal.

5. The method of claim 1, further comprising using at least a portion of the buffered input signal to synchronize a timing of audio data with a timing of the input signal.

6. The method of claim 1, further comprising detecting an indication to stop receiving the audio data and responsively stopping receipt of the audio data, and starting the buffering of the input signal.

7. The method of claim 1, wherein receiving the audio data via the one or more computer networks includes receiving first audio data of the audio data wirelessly over one or more radio frequency channels.

8. The method of claim 1, wherein the receiving of the audio data comprises receiving first audio data of the audio data from a source of a first interference component of the one or more interference components, and receiving second audio data of the audio data from a source of a second interference component of the one or more interference components.

9. The method of claim 1, wherein the providing of the output signal for speech recognition comprises transmitting the output signal to a speech recognition application communicatively coupled to a first computer network of the one or more computer networks.

10. A voice controlled hub comprising:

a microphone array configured to capture a voice command of a human user and speaker sound generated by a remote speaker system and provide an audio input signal comprising a Voice command component and a speaker sound component;

a radio frequency transceiver configured to receive, from the remote speaker system, stream data including audio data that was used by the remote speaker system to generate the speaker sound;

a buffer configured to buffer the audio input signal at least until e radio frequency transceiver receives the stream data from the remote speaker system; and

a processing system configured to use a noise canceller to distinguish between the voice command component of the audio input signal and the speaker sound component of the audio input signal, wherein the noise canceller is configured to use the stream data to remove at least a portion of the speaker sound component from the audio input signal to generate an estimated voice command component.

11. The voice controlled hub of claim 10, further comprising a packet loss concealer, wherein the processing system is configured to use the packet loss concealer to conceal packets of the stream data that are unavailable for use by the noise canceller.

12. The voice controlled hub of claim 10, further comprising a memory system, wherein the processing system is configured to use the memory system to buffer the audio input signal to synchronize a timing of the stream data with a timing of the audio input signal.

13. The voice controlled hub of claim 12, wherein the processing system is configured to detect a voice activation of the voice controlled hub and responsively use the radio frequency transceiver to request the stream data from the remote speaker system.

14. The voice controlled hub of claim 10, wherein the radio frequency transceiver is configured to transmit the estimated voice command component to a remote automatic speech recognition application and receive voice command data corresponding to the estimated voice command, and wherein the processing system is configured to initiate a device function based on the voice command data.