Systems and methods for determining microphone position

• US 10,085,101 B2
• Filed: 07/13/2016
• Issued: 09/25/2018
• Est. Priority Date: 07/13/2016
• Status: Active Grant

• Current Assignee: Hand Held Products Incorporated (Honeywell International Inc.)
• Original Assignee: Hand Held Products Incorporated (Honeywell International Inc.)
• Inventors: Hardek, David D.
• Primary Examiner(s): Sniezek, Andrew L
• Application Number:
  US15/209,145
• Publication Number:
  US 20180020305A1
• Time in Patent Office: 804 Days
• CPC Class Codes: G10L 15/01 Assessment or evaluation of speech recognition systems
• CPC Class Codes: G10L 15/16 using artificial neural networks
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence
• CPC Class Codes: G10L 2015/025 Phonemes, fenemes or fenones being the recognition units
• CPC Class Codes: G10L 2015/0631 Creating reference templates; Clustering
• CPC Class Codes: H04R 2201/107 Monophonic and stereophonic headphones with microphone for two-way hands free communication
• CPC Class Codes: H04R 29/004 for microphones H04R29/007 takes precedence
• CPC Class Codes: H04R 5/033 Headphones for stereophonic communication details thereof, e.g. relating to batteries, cables or control elements H04R1/10

Abstract

A method for determining a relative position of a microphone may include capturing speech audio from a user'"'"'s mouth with the microphone so that the microphone outputs an electrical signal indicative of the speech audio; determining an indication of a position of the microphone relative to the user'"'"'s mouth, which may include providing a plurality of inputs to a computerized discriminative classifier, wherein an input of the plurality of inputs is derived from the electrical signal, and wherein an output from the computerized discriminative classifier is indicative of the position of the microphone relative to the user'"'"'s mouth.

Claims

1. A method for determining a relative position of a microphone, the method comprising:

capturing speech audio from a user'"'"'s mouth with the microphone so that the microphone outputs an electrical signal indicative of the speech audio;

a computer deriving a plurality of inputs from the electrical signal;

determining a derived value of an approximate position of the microphone relative to the user'"'"'s mouth comprising providing, to a discriminative classifier implemented on the computer;

at least the plurality of inputs and at least some contextual data, the contextual data originating and/or representing conditions occurring at a time when the speech audio is captured;

the discriminative classifier comprising a model derived from training data and/or test data, the training data and/or the test data comprising a manually measured actual position of a microphone relative to a user'"'"'s mouth;

the computer receiving an output from the discriminative classifier, the output providing the derived value of the approximate position of the microphone relative to the user'"'"'s mouth based at least in part on the plurality of inputs and the contextual data; and

the computer determining whether the derived value of the approximate position of the microphone is unacceptable at least in part by comparing the derived value to a value or range indicative of the microphone being an acceptable distance relative to the user'"'"'s mouth, and providing a signal to a user if the derived value of the approximate position of the microphone is unacceptable.

2. The method of claim 1, comprising the computer calculating a Fourier transformation on data selected from the group consisting of the electrical signal and data derived from the electrical signal.

3. The method of claim 2, wherein an input of the plurality of inputs is derived from results from the calculating of the Fourier transformation.

4. The method of claim 1, comprising the computer decoding a phoneme from data selected from the group consisting of the electrical signal and data derived from the electrical signal, wherein an input of the plurality of inputs comprises the phoneme.

5. The method of claim 4, comprising the computer decoding the phoneme using a text-to-phoneme engine.

6. The method of claim 1, comprising:

deriving first and second inputs of the plurality of inputs from the electrical signal; and

weighting the first input more heavily than any weighting of the second input in the discriminative classifier.

7. The method of claim 6, comprising providing first and second phenomes that are different from one another, comprising performing text-to-phenome conversions, wherein:

the first input comprises the first phenome; and

the second input comprises the second phenome.

8. The method of claim 1, wherein the contextual data comprises at least one of a gain setting, and/or a classification of background noise.

9. A method for determining a relative position of a microphone, the method comprising:

capturing speech audio from a user'"'"'s mouth with the microphone so that the microphone outputs an electrical signal indicative of the speech audio;

determining a derived value of an approximate position of the microphone relative to the user'"'"'s mouth, comprising providing a plurality of inputs to a computerized discriminative classifier, the discriminative classifier comprising a model derived from training data and/or test data, the training data and/or the test data comprising a manually measured actual position of a microphone relative to a user'"'"'s mouth wherein;

a first input of the plurality of inputs is derived from the electrical signal, and a second input of the plurality of inputs comprises contextual data, the contextual data originating and/or representing conditions occurring at a time when the speech audio is captured; and

an output from the computerized discriminative classifier is the derived value of the approximate position of the microphone relative to the user'"'"'s mouth, the output derived at least in part from the plurality of inputs.

10. The method of claim 9, comprising:

a computer determining whether the derived value of the approximate position of the microphone is unacceptable; and

the computer providing a signal in response to the computer determining that the derived value of the approximate position of the microphone is unacceptable.

11. The method of claim 9, comprising a computer deriving the input from the electrical signal.

12. The method of claim 11, comprising calculating a Fourier transformation on data selected from the group consisting of the electrical signal and data derived from the electrical signal.

13. The method of claim 12, wherein the input comprises results from the calculating of the Fourier transformation.

14. The method of claim 12, wherein the input is derived from results from the calculating of the Fourier transformation.

15. The method of claim 11, comprising decoding a phoneme from data selected from the group consisting of the electrical signal and data derived from the electrical signal.

16. The method of claim 15, wherein the input comprises the phoneme, and the decoding of the phoneme is comprised of using a text-to-phoneme engine.

17. The method of claim 9, comprising:

deriving first and second inputs of the plurality of inputs from the electrical signal; and

weighting the first input more heavily than any weighting of the second input in the computerized discriminative classifier.

18. The method of claim 17, comprising providing first and second phenomes that are different from one another, comprising performing text-to-phenome conversions, wherein:

the first input comprises the first phenome; and

the second input comprises the second phenome.

19. A method for determining a relative position of a microphone, the method comprising:

providing a plurality of inputs to a discriminative classifier implemented on a computer, the discriminative classifier comprising a model derived from training data and/or test data, the training data and/or the test data comprising a manually measured actual position of a microphone relative to a user'"'"'s mouth, and the plurality of inputs comprising;

(i) an electrical signal output from the microphone in response to the microphone capturing speech audio from a user'"'"'s mouth while the microphone is at a position relative to the user'"'"'s mouth, and/or data derived from the electrical signal; and

(ii) contextual data, the contextual data originating and/or representing conditions occurring at a time when the speech audio is captured;

the computer receiving an output from the discriminative classifier, the output providing a derived value of an approximate position of the microphone relative to the user'"'"'s mouth, the derived value based at least in part on the plurality of inputs; and

the computer determining whether the derived value of the approximate position of the microphone is unacceptable at least in part by comparing the derived value output from the discriminative classifier to a value or range indicative of the microphone being an acceptable distance relative to the user'"'"'s mouth, and providing a signal if the derived value of the approximate position of the microphone is unacceptable.

20. The method of claim 19, wherein the microphone is part of a head set that further comprises a speaker, and the method comprises the speaker providing an audio indication that the position of the microphone is unacceptable, wherein the speaker providing the audio indication is in response to the computer providing the signal.

21. The method of claim 19, comprising deriving the input from the electrical signal, wherein the input is selected from the group consisting of a Fourier transform and a phenome.