Method and system for preventing speech comprehension by interactive voice response systems

• US 7,979,274 B2
• Filed: 05/20/2009
• Issued: 07/12/2011
• Est. Priority Date: 10/01/2004
• Status: Expired due to Fees

• Current Assignee: Nuance Communications, Inc. (Microsoft Corporation)
• Original Assignee: AT&T Intellectual Property I LP (AT&T, Inc.)
• Inventors: DeSimone, Joseph
• Primary Examiner(s): YEN, ERIC L
• Application Number:
  US12/469,106
• Publication Number:
  US 20090228271A1
• Time in Patent Office: 783 Days
• Field of Search: 704/226, 704/231
• US Class Current: 704/226
• CPC Class Codes: G10L 13/10 Prosody rules derived from text; Stress or intonation

Abstract

A method of and system for generating a speech signal with an overlayed random frequency signal using prosody modification of a speech signal output by a text-to-speech (TTS) system to substantially prevent an interactive voice response (IVR) system from understanding the speech signal without significantly degrading the speech signal with respect to human understanding. The present invention involves modifying a prosody of the speech output signal by using a prosody of the user'"'"'s response to a prompt. In addition, a randomly generated overlay frequency is used to modify the speech signal to further prevent the IVR system from recognizing the TTS output. The randomly generated frequency may be periodically changed using an overlay timer that changes the random frequency signal at a predetermined intervals.

Claims

1. A method of modifying a speech signal for reducing the likelihood for recognition of the speech signal by a speech recognition system, the method comprising:

receiving at least one prosody sample; and

modifying at least one prosody characteristic of an initial speech signal based on the at least one prosody sample, thereby generating a modified speech signal, the modified speech signal being less likely to be recognized by a speech recognition system than the initial speech signal, wherein the modified speech signal is further altered by;

(a) obtaining an acceptable frequency range;

(b) calculating a random frequency signal;

(c) comparing the random frequency signal to the acceptable frequency range;

(d) repeating steps (b) and (c) in response to the calculated random frequency signal not being within the acceptable frequency range; and

(e) overlaying the random frequency signal onto the modified speech signal in response to the random frequency signal being within the acceptable frequency range.

2. A method of modifying a speech signal as defined in claim 1, further comprises:

prompting a user; and

wherein the at least one prosody sample is received from the user in response to the prompting.

3. A method of modifying a speech signal as defined in claim 1, further comprises:

generating a random frequency signal; and

overlaying the random frequency signal on the modified speech signal.

4. A method of modifying a speech signal as defined in claim 1, further comprising:

initializing an overlay timer, the overlay timer being adapted to expire at a predetermined time;

determining if the overlay timer has expired;

generating the modified speech signal in response to the overlay timer not having expired; and

recalculating the random frequency signal in response to the initial overlay timer expiring.

5. A method of modifying a speech signal as defined in claim 4, further comprises:

(a) obtaining a first random number;

(b) measuring a variable parameter;

(c) equating a second number to the variable parameter;

(d) dividing the first random number by the second number to generate a quotient;

(e) determining whether the quotient is within numeric values defined by the acceptable frequency range;

(f) performing steps (a)-(d) until the quotient is within the acceptable frequency range; and

(g) equating the quotient to the random frequency signal in response to the quotient being within the acceptable frequency range.

6. A method of modifying a speech signal as defined in claim 5, wherein the second random number comprises the measured outside ambient temperature.

7. A method of modifying a speech signal as defined in claim 5, wherein the second random number comprises the outside wind speed.

8. A method of modifying a speech signal as defined in claim 7, wherein the resultant random frequency signal number is rounded to the fifth decimal place.

9. A method of modifying a speech signal as defined in claim 1, wherein the acceptable frequency range is within the audible human hearing range.

10. A method of modifying a speech signal as defined in claim 9, wherein the acceptable frequency range is between 20 Hz and 8,000 Hz.

11. A method of modifying a speech signal as defined in claim 9, wherein the acceptable frequency range is between 16,000 Hz and 20,000 Hz.

12. A method of modifying a speech signal for reducing the likelihood of recognition of the speech signal by a speech recognition system, the method comprising:

accessing a text file;

utilizing a text-to-speech synthesizer to generate a speech signal from the text file;

receiving a prosody sample from a user in response to prompting; and

modifying the speech signal with a characteristic of the prosody sample such that an audio output of the modified speech signal is less likely to be understood by a speech recognition system than an audible output of the generated speech signal, wherein the modified speech signal is further altered by;

(a) obtaining an acceptable frequency range;

(b) calculating a random frequency signal;

(c) comparing the random frequency signal to the acceptable frequency range;

(d) repeating steps (b) and (c) in response to the calculated random frequency signal not being within the acceptable frequency range; and

(e) overlaying the random frequency signal onto the modified speech signal in response to the random frequency signal being within the acceptable frequency range.

13. A method of modifying a speech signal as defined in claim 12, further comprises:

generating a random frequency signal; and

overlaying the random frequency signal on the modified speech signal.

14. A method of modifying a speech signal as defined in claim 12, further comprising:

initializing an overlay timer, the overlay timer being adapted to expire at a predetermined time;

determining if the overlay timer has expired;

generating the modified speech signal in response to the overlay time not having expired; and

recalculating the random frequency signal in response to the overlay timer expiring.

15. A method of modifying a speech signal as defined in claim 14, further comprises:

(a) obtaining a first random number;

(b) measuring a variable parameter;

(c) equating a second number to the variable parameter;

(d) dividing the first random number by the second number to generate a quotient;

(e) determining whether the quotient is within numeric values defined by an acceptable frequency range;

(f) performing steps (a)-(d) until the quotient is within the acceptable frequency range; and

(g) equating the quotient to the random frequency signal in response to the quotient being within the acceptable frequency range.

16. A method of modifying a speech signal defined in claim 15, wherein the second random number comprises the measured outside ambient temperature.

17. A method of modifying a speech signal as defined in claim 15, wherein the second random number comprises the outside wind speed.

18. A method of modifying a speech signal as defined in claim 17, wherein the resultant random frequency signal number is rounded to the fifth decimal place.

19. A method of modifying a speech signal as defined in claim 12, wherein the acceptable frequency range is within the audible human hearing range.

20. A method of modifying a speech signal as defined in claim 19, wherein the acceptable frequency range is between 20 Hz and 8,000 Hz.

21. A method of modifying a speech signal as defined in claim 19, wherein the acceptable frequency range is between 16,000 Hz and 20,000 Hz.

22. A system for decreasing the likelihood of recognition of a speech signal by a speech recognition system, the system comprising:

a receiver for receiving at least one prosody sample; and

a speech signal modifier modifying at least one prosody characteristic associated with an initial speech signal in accordance with the at least one prosody sample, thereby generating a modified speech signal, the modified speech signal being less likely to be recognized by a speech recognition system than the initial speech signal, wherein the modified speech signal is further altered by;

(a) obtaining an acceptable frequency range;

(b) calculating a random frequency signal;

(c) comparing the random frequency signal to the acceptable frequency range;

(d) repeating steps (b) and (c) in response to the calculated random frequency signal not being within the acceptable frequency range; and

(e) overlaying the random frequency signal onto the modified speech signal in response to the random frequency signal being within the acceptable frequency range.

23. A system for decreasing the recognition of a speech signal by a speech recognition system as defined in claim 22, further comprising a frequency overlay subsystem, the frequency overlay subsystem generating a random frequency signal to overlay on the modified speech signal.

24. A system for decreasing the recognition of a speech signal by a speech recognition system as defined in claim 23, wherein the frequency overlay subsystem further comprises an overlay timer being adapted to expire at a predetermined time to indicate the generation of a random frequency.