Microphone setup and testing in voice recognition software

• US 7,243,068 B2
• Filed: 09/10/2004
• Issued: 07/10/2007
• Est. Priority Date: 09/10/2004
• Status: Active Grant

• Current Assignee: Scientific Learning Corp. (Carnegie Learning, Inc.)
• Original Assignee: Soliloquy Learning, Inc. (Carnegie Learning, Inc.)
• Inventors: Barker, Simon, Wolf, Jared J.
• Primary Examiner(s): Lerner; Martin
• Application Number:
  US10/939,167
• Publication Number:
  US 20060069557A1
• Time in Patent Office: 1,033 Days
• Field of Search: 704/226, 704/227, 704/228, 704/231, 704/233, 381/57, 381/58, 700/94
• US Class Current: 704/233
• CPC Class Codes: G10L 15/01 Assessment or evaluation of speech recognition systems

Abstract

Systems, devices, and methods for performing a microphone test routine are disclosed during an initial setup of a speech recognition program and at a predetermined time intervals during use of the speech recognition program to detect recognition errors associated with the microphone.

Claims

1. A method comprising:

performing a microphone test routine during an initial setup of a speech recognition program, the routine including;

analyzing signal quality of a received electrical signal from a microphone; and

performing, at a predetermined time intervals, the microphone testing routine during use of the speech recognition program to detect recognition errors associated with the microphone; and

signaling a problem associated with the microphone after a predetermined number of errors associated with the microphone have been detected, wherein the errors associated with the microphone include errors associated with wind noise in the microphone caused by a user'"'"'s breath stream.

2. The method of claim 1 wherein the errors associated with the microphone include errors associated with hum in the signal received from the microphone.

3. The method of claim 1 wherein the errors associated with the microphone include errors associated with a low signal to noise ratio (SNR).

4. The method of claim 1 wherein the errors associated with the microphone include errors associated with clipping of audio received by the microphone.

5. The method of claim 1 wherein the errors associated with the microphone include errors associated with a loss of connection to the microphone.

6. The method of claim 1 further comprising automatically adjusting a parameter for the microphone based on the signal quality.

7. The method of claim 1 further comprising tracking the errors.

8. The method of claim 1 further comprising setting a gain for the microphone.

9. A method for testing a signal quality for a microphone, the method comprising:

testing a first signal to determine if hum exists in the first signal;

testing a second signal to determine if breath pops are present in the second signal;

testing a third signal to determine if a signal to noise ratio for the third signal is within an acceptable range.

10. The method of claim 9 wherein testing a signal to determine if breath pops are present in the second signal includes:

analyzing a test phrase including words beginning with at least one of the letter ‘

p’ and

the letter ‘

t’

.

11. The method of claim 10 wherein analyzing a test phrase includes:

determining speech energy attributed to speech in the received signal corresponding to the test phase;

determining noise energy attributed to noise in the received signal; and

comparing the speech energy to the noise energy.

12. The method of claim 10 wherein analyzing a test phrase includes analyzing a signal in a particular frequency range.

13. The method of claim 12 wherein the frequency range is in a range about a frequency of nearby a.c. power sources that may interfere with the signal from the microphone.

14. The method of claim 12 wherein the frequency range is between approximately 55 Hz and 65 Hz.

15. The method of claim 9 wherein testing a signal to determine if a hum exists includes detecting presence of an approximately constant signal component.

16. The method of claim 15 wherein the approximately constant signal is within a predetermined frequency range.

17. The method of claim 16 wherein the predetermined frequency range is approximately 55-65 Hz.

18. The method of claim 9 further comprising testing a sound level of the signal.

19. The method of claim 9 wherein the first signal, second signal, and third signal are the same signal.

20. The method of claim 9 wherein the first signal, second signal, and third signal are each different signals.

21. A computer program product residing on a computer readable medium comprising instructions for causing an electrical device to:

perform a microphone test during an initial setup of a speech recognition program, the routine including instructions to;

analyze signal quality of a received electrical signal from a microphone;

set a gain for the microphone;

perform at a predetermined regular time interval the microphone testing routine during use of the speech recognition program to detect recognition errors associated with the microphone; and

signal a problem associated with the microphone after a predetermined number of errors associated with the microphone have been detected, wherein the routine includes instructions to detect errors associated with wind noise in the microphone caused by a user'"'"'s breath stream.

22. The computer program of claim 21 wherein the routine includes instructions to detect errors associated with hum in the signal received from the microphone.

23. The computer program of claim 21 wherein the routine includes instructions to detect errors associated with a low signal to noise ratio (SNR).

24. A computer program product residing on a computer readable medium for testing a signal quality for a microphone, comprising instructions for causing an electrical device to:

test a first signal to determine if a hum exists in the first signal;

test a second signal to determine if breath pops are present in the second signal; and

test a third signal to determine if a signal to noise ratio for the third signal is within an acceptable range.

25. The computer program of claim 24 wherein instructions to test a signal to determine if breath pops are present in the second signal includes instructions to:

analyze a test phrase including words beginning with at least one of the letter ‘

p’ and

the letter ‘

t’

.

26. The computer program of claim 25 wherein instructions to analyze includes instructions to analyze a signal in a particular frequency range.

27. The computer program of claim 24 wherein instructions to analyze include:

determine speech energy attributed to speech in the received signal corresponding to the test phrase;

determine noise energy attributed to noise in the received signal; and

compare the speech energy to the noise energy.

28. A device configured to:

perform a microphone test during an initial setup of a speech recognition program, the routine including instructions to;

analyze signal quality of a received electrical signal from a microphone;

set a gain for the microphone;

perform at a predetermined regular time interval the microphone testing routine during use of the speech recognition program to detect recognition errors associated with the microphone; and

signal a problem associated with the microphone after a predetermined number of errors associated with the microphone have been detected, wherein the routine includes instructions to detect errors associated with wind noise in the microphone caused by a user'"'"'s breath stream.

29. The device of claim 28 wherein the routine includes instructions to detect errors associated with hum in the signal received from the microphone.

30. The device of claim 28 wherein the routine includes instructions to detect errors associated with a low signal to noise ratio (SNR).

31. The device of claim 28 wherein instructions to analyze include:

determine speech energy attributed to speech in the received signal corresponding to the test phrase;

determine noise energy attributed to noise in the received signal; and

compare the speech energy to the noise energy.

32. A device configured to:

test a first signal to determine if a hum exists in the first signal;

test a second signal to determine if breath pops are present in the second signal; and

test a third signal to determine if a signal to noise ratio for the third signal is within an acceptable range.

33. The device of claim 32 wherein instructions to test a signal to determine if breath pops are present in the second signal includes instructions to:

analyze a test phrase including words beginning with at least one of the letter ‘

p’ and

the letter ‘

t’

.

34. The device of claim 33 wherein instructions to analyze includes instructions to analyze a signal in a particular frequency range.

35. A method comprising:

performing a microphone test routine during an initial setup of a speech recognition program, the routine including;

analyzing signal quality of a received electrical signal from a microphone; and

performing, at a predetermined time intervals, the microphone testing routine during use of the speech recognition program to detect recognition errors associated with the microphone; and

signaling a problem associated with the microphone after a predetermined number of errors associated with the microphone have been detected, wherein the errors associated with the microphone include errors associated with hum in the signal received from the microphone.

36. The method of claim 35 wherein the errors associated with the microphone include errors associated with a low signal to noise ratio (SNR).

37. The method of claim 35 wherein the errors associated with the microphone include errors associated with clipping of audio received by the microphone.

38. The method of claim 35 wherein the errors associated with the microphone include errors associated with a loss of connection to the microphone.

39. A computer program product residing on a computer readable medium comprising instructions for causing an electrical device to:

perform a microphone test during an initial setup of a speech recognition program, the routine including instructions to;

analyze signal quality of a received electrical signal from a microphone;

set a gain for the microphone;

perform at a predetermined regular time interval the microphone testing routine during use of the speech recognition program to detect recognition errors associated with the microphone; and

signal a problem associated with the microphone after a predetermined number of errors associated with the microphone have been detected, wherein the routine includes instructions to detect errors associated with hum in the signal received from the microphone.

40. The computer program of claim 39 wherein the routine includes instructions to detect errors associated with a low signal to noise ratio (SNR).

41. A device configured to:

perform a microphone test during an initial setup of a speech recognition program, the routine including instructions to;

analyze signal quality of a received electrical signal from a microphone;

set a gain for the microphone;

perform at a predetermined regular time interval the microphone testing routine during use of the speech recognition program to detect recognition errors associated with the microphone; and

signal a problem associated with the microphone after a predetermined number of errors associated with the microphone have been detected, wherein the routine includes instructions to detect errors associated with hum in the signal received from the microphone.

42. The device of claim 41 wherein the routine includes instructions to detect errors associated with a low signal to noise ratio (SNR).