Device and method for adjusting speech intelligibility at an audio device

• US 10,431,237 B2
• Filed: 09/13/2017
• Issued: 10/01/2019
• Est. Priority Date: 09/13/2017
• Status: Active Grant

• Current Assignee: Motorola Solutions, Inc.
• Original Assignee: Motorola Solutions, Inc.
• Inventors: Mohan, Glenn Andrew, Howell, Maurice D., Giol, Juan J., Ibarra, Christian
• Primary Examiner(s): Chawan, Vijay B
• Application Number:
  US15/702,815
• Publication Number:
  US 20190080708A1
• Time in Patent Office: 748 Days
• Field of Search: 704233, 704226-227, 704201, 704270, 7042701, 704275
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue
• CPC Class Codes: G10L 2021/03646 Stress or Lombard effect
• 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 25/51 for comparison or discrimination
• CPC Class Codes: G10L 25/60 for measuring the quality of voice signals
• CPC Class Codes: G10L 25/84 for discriminating voice from noise

Abstract

A device and method for adjusting speech intelligibility at an audio device is provided. The device comprises a microphone, a transmitter and a controller. The controller is configured to: determine a noise level at the microphone; select a voice tag, of a plurality of voice tags, based on the noise level, each of the plurality of voice tags associated with respective noise levels; determine an intelligibility rating of a mix of the voice tag and noise received at the microphone; and when the intelligibility rating is below a threshold intelligibility rating, enhance speech received the microphone based on the intelligibility rating prior to transmitting, at the transmitter, a signal representing intelligibility enhanced speech.

Claims

1. A device comprising:

a microphone;

a transmitter; and

a controller having access to a memory storing a plurality of preconfigured voice tags associated with respective noise levels, each of the plurality of preconfigured voice tags comprising a respective voice recording of a given user,

the controller configured to;

determine a noise level at the microphone;

select a voice tag, of the plurality of preconfigured voice tags, based on the noise-level;

determine an intelligibility rating of a mix of the voice tag and noise received at the microphone; and

when the intelligibility rating is below a threshold intelligibility rating, enhance speech from the given user received at the microphone based on the intelligibility rating prior to transmitting, at the transmitter, a signal representing intelligibility enhanced speech.

2. The device of claim 1, wherein each of respective voice recordings of the plurality of preconfigured voice tags are associated with a respective Lombard Speech Level.

3. The device of claim 1, wherein the controller is further configured to, prior to determining the intelligibility rating of the mix:

determine a signal-to-noise ratio of the speech received at the microphone;

match the mix to the signal-to-noise ratio of the speech received at the microphone; and

enhance the mix of the voice tag and the noise using a baseline speech enhancement filter.

4. The device of claim 1, wherein the controller is further configured to determine the intelligibility rating of the mix of the voice tag and the noise received at the microphone by:

comparing the mix, enhanced using a baseline speech enhancement filter, with the voice tag.

5. The device of claim 1, wherein the controller is further configured to determine the intelligibility rating of the mix by:

binning the mix based on frequency;

determining respective intelligibility ratings for a plurality of bins; and

averaging the respective intelligibility ratings.

6. The device of claim 5, wherein the averaging of the respective intelligibility rating comprises a weighted averaging.

7. The device of claim 1, wherein the controller is further configured to determine the intelligibility rating of the mix by determining one or more of:

amplitude modulation at different frequencies in the mix;

speech presence or speech absence at different frequencies in the mix;

respective noise levels at the different frequencies;

respective reverberation at the different frequencies;

respective signal-to-noise ratio at the different frequencies;

speech coherence at the different frequencies; and

speech distortion at the different frequencies.

8. The device of claim 1, wherein the controller is further configured to enhance the speech received at the microphone based on the intelligibility rating using one or more of noise suppression, speech reconstruction and an equalizer.

9. The device of claim 1, wherein the controller is further configured to:

continue to determine the intelligibility rating of further speech received at the microphone; and

when the intelligibility rating remains below the threshold intelligibility rating, control a notification device to prompt a user to change one or more of a speaking behavior and a position of the microphone.

19. The device of claim 1, wherein each of respective voice recordings of the plurality of preconfigured voice tags are recorded at different speaking levels associated with an associated respective noise level.

10. A method comprising:

determining, at a controller of a device, a noise level at a microphone of the device;

selecting, at the controller, a voice tag, of a plurality of preconfigured voice tags, based on the noise level, each of the plurality of preconfigured voice tags associated with respective noise levels, the plurality of preconfigured voice tags stored at a memory accessible to the controller, each of the plurality of preconfigured voice tags comprising a respective voice recording of a given user;

determining, at the controller, an intelligibility rating of a mix of the voice tag and noise received at the microphone; and

when the intelligibility rating is below a threshold intelligibility rating, enhancing, using the controller, speech of the given user received the microphone based on the intelligibility rating prior to transmitting, at a transmitter of the device, a signal representing intelligibility enhanced speech.

11. The method of claim 10, wherein each of respective voice recordings of the plurality of preconfigured voice tags are associated with a respective Lombard Speech Level.

12. The method of claim 10, further comprising, prior to determining the intelligibility rating of the mix:

determining, at the controller, a signal-to-noise ratio of the speech received at the microphone;

matching, at the controller, the mix to the signal-to-noise ratio of the speech received at the microphone; and

enhancing, at the controller, the mix of the voice tag and the noise using a baseline speech enhancement filter.

13. The method of claim 10, further comprising determining the intelligibility rating of the mix of the voice tag and the noise received at the microphone by:

comparing the mix, enhanced using a baseline speech enhancement filter, with the voice tag.

14. The method of claim 10, wherein the controller is further configured to determine the intelligibility rating of the mix by:

binning, at the controller, the mix based on frequency;

determining, at the controller, respective intelligibility ratings for a plurality of bins; and

averaging, at the controller, the respective intelligibility ratings.

15. The method of claim 14, wherein the averaging of the respective intelligibility rating comprises a weighted averaging.

16. The method of claim 10, further comprising determining the intelligibility rating of the mix by determining, at the controller, one or more of:

amplitude modulation at different frequencies in the mix;

speech presence or speech absence at different frequencies in the mix;

respective noise levels at the different frequencies;

respective reverberation at the different frequencies;

respective signal-to-noise ratio at the different frequencies;

speech coherence at the different frequencies; and

speech distortion at the different frequencies.

17. The method of claim 10, further comprising enhancing, at the controller, the speech received at the microphone based on the intelligibility rating using one or more of noise suppression, speech reconstruction and an equalizer.

18. The method of claim 10, further comprising:

continuing to determine, at the controller, the intelligibility rating of further speech received at the microphone; and

when the intelligibility rating remains below the threshold intelligibility rating, controlling, at the controller, a notification device to prompt a user to change one or more of a speaking behavior and a position of the microphone.

20. The method of claim 10, wherein each of respective voice recordings of the plurality of preconfigured voice tags are recorded at different speaking levels associated with an associated respective noise level.