Apparatus and Method for Power Efficient Signal Conditioning for a Voice Recognition System

• US 20140278393A1
• Filed: 07/31/2013
• Published: 09/18/2014
• Est. Priority Date: 03/12/2013
• Status: Abandoned Application

• Current Assignee: Google Technology Holdings LLC (Alphabet Inc.)
• Original Assignee: Motorola Mobility LLC (Lenovo Group Ltd.)
• Inventors: Ivanov, Plamen A., Clark, Joel A., Jasiuk, Mark A., Ramabadran, Tenkasi V., Wu, Jincheng, Bastry, Kevin J.
• Application Number:
  US13/955,186
• Publication Number:
  US 20140278393A1
• Time in Patent Office: Days
• US Class Current: 704/233
• 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 15/28 Constructional details of speech recognition systems
• CPC Class Codes: G10L 2021/02161 Number of inputs available containing the signal or the noise to be suppressed
• CPC Class Codes: G10L 2025/783 based on threshold decision
• CPC Class Codes: G10L 21/02 Speech enhancement, e.g. noise reduction or echo cancellation reducing echo effects in line transmission systems H04B3/20; echo suppression in hands-free telephones H04M9/08
• CPC Class Codes: G10L 21/0216 characterised by the method used for estimating noise
• CPC Class Codes: G10L 21/0272 Voice signal separating
• CPC Class Codes: G10L 21/0364 for improving intelligibility
• CPC Class Codes: G10L 25/21 the extracted parameters being power information
• CPC Class Codes: G10L 25/78 Detection of presence or absence of voice signals switching of direction of transmission by voice frequency in two-way loud-speaking telephone systems H04M9/10
• CPC Class Codes: G10L 25/84 for discriminating voice from noise

Abstract

A disclosed method includes monitoring an audio signal energy level while having a plurality of signal processing components deactivated and activating at least one signal processing component in response to a detected change in the audio signal energy level. The method may include activating and running a voice activity detector on the audio signal in response to the detected change where the voice activity detector is the at least one signal processing component. The method may further include activating and running the noise suppressor only if a noise estimator determines that noise suppression is required. The method may activate and runs a noise type classifier to determine the noise type based on information received from the noise estimator and may select a noise suppressor algorithm, from a group of available noise suppressor algorithms, where the selected noise suppressor algorithm is the most power consumption efficient.

Claims

1. A method comprising:

monitoring an audio signal energy level with a plurality of signal processing components deactivated; and

activating at least one signal processing component of the plurality of signal processing components in response to a detected change in the audio signal energy level.

2. The method of claim 1, wherein activating at least one signal processing component comprises:

activating and running a voice activity detector on the audio signal in response to the detected change in the audio energy level, the voice activity detector being one of the plurality of signal processing components.

3. The method of claim 2, further comprising:

activating and running a noise estimator in response to voice being detected in the audio signal by the voice activity detector.

4. The method of claim 3, further comprising:

determining, by the noise estimator, that noise suppression is required for the audio signal; and

activating and running a noise suppressor on the audio signal in response to the noise estimator determination.

5. The method of claim 3, further comprising:

activating and running a noise type classifier to determine a noise type based on information received from the noise estimator; and

selecting a noise suppressor algorithm based on the determined noise type.

6. The method of claim 3, further comprising:

determining, by the noise estimator, that noise suppression is not required for the audio signal; and

performing voice recognition on the audio signal without activating a noise suppressor.

7. The method of claim 1, further comprising:

activating at least one additional microphone to receive the audio signal in response to the detected change in the audio signal energy level.

8. The method of claim 7, further comprising:

deactivating the at least one additional microphone and returning to a single microphone configuration in response to voice not being detected in the audio signal by a voice activity detector or a second detected change in the audio signal energy level.

9. The method of claim 1, further comprising:

calculating, by an energy estimator, a long term energy baseline and a short term deviation wherein monitoring the audio signal energy level is performed by the energy estimator.

10. The method of claim 9, further comprising:

buffering the audio signal in response to a detected short term deviation.

11. An apparatus comprising:

a noise suppressor;

a voice activity detector; and

an energy estimator, operatively coupled to the voice activity detector, the energy estimator operative to monitor an audio signal energy level with at least the noise suppressor and the voice activity detector deactivated and to activate at least the voice activity detector in response to a detected change in the audio signal energy level.

12. The apparatus of claim 11, further comprising:

a noise estimator, operatively coupled to the voice activity detector, wherein the voice activity detector is operative to activate the noise estimator in response to voice being detected in the audio signal.

13. The apparatus of claim 12, further comprising:

a buffer, operatively coupled to the energy estimator, the buffer operative to receive a buffer control signal from the energy estimator and to buffer the audio signal in response to the buffer control signal, the energy estimator operative to send the buffer control signal in response to the detected change in the audio signal energy level.

14. The apparatus of claim 13, further comprising:

a switch, operatively coupled to the noise suppressor and to the noise estimator to receive a switch control signal, the switch operative to change over between a noise suppressed audio signal output from the noise suppressor and a buffered audio signal output from the buffer, according to the switch control signal; and

wherein the noise estimator is operative to send the switch control signal.

15. The apparatus of claim 14, further comprising:

a noise suppressor algorithms selector, operatively coupled to the noise estimator and to the noise suppressor, the noise suppressor algorithms selector operative to activate and run the noise suppressor in response to a noise estimator control signal sent when the noise estimator determines that noise suppression is required.

16. The apparatus of claim 15, further comprising:

a noise type classifier, operatively coupled to the noise estimator and to the noise suppressor algorithms selector, the noise type classifier operative to activate and run in response to a control signal from the noise estimator, and operative to determine noise type based on information received from the noise estimator.

17. The apparatus of claim 16, wherein the noise suppressor algorithms selector is further operative to select a noise suppressor algorithm based on the noise type determined by the noise type classifier.

18. The apparatus of claim 14, where the noise estimator is further operative to determine that noise suppression is not required and send the switch control signal to change over from the noise suppressed audio signal output from the noise suppressor to the buffered audio signal output from the buffer.

19. The apparatus of claim 11, further comprising:

a plurality of microphones; and

microphone configuration logic operative to turn each microphone on or off; and

wherein the energy estimator is further operative to control the microphone configuration logic to turn on at least one additional microphone in response to the detected change in the audio signal energy level.

20. The apparatus of claim 19, wherein the voice activity detector is operative to deactivate the at least one additional microphone and return to a single microphone configuration in response to voice not being detected in the audio signal by the voice activity detector.

21. The apparatus of claim 14, further comprising:

voice command recognition logic, having an input operatively coupled to the switch.

22. The apparatus of claim 21, further comprising:

a transceiver having an input operatively coupled to the switch.