Biometric password security

• US 9,437,195 B2
• Filed: 09/18/2013
• Issued: 09/06/2016
• Est. Priority Date: 09/18/2013
• Status: Active Grant

• Current Assignee: Lenovo PC International Limited (Lenovo Group Ltd.)
• Original Assignee: Lenovo Singapore Pte Limited (Lenovo Group Ltd.)
• Inventors: Nicholson, John Weldon, Perrin, Steven Richard
• Primary Examiner(s): Saint Cyr, Leonard
• Application Number:
  US14/030,919
• Publication Number:
  US 20150081301A1
• Time in Patent Office: 1,084 Days
• Field of Search: 704246-252
• US Class Current: 1/1
• CPC Class Codes: G10L 17/24 the user being prompted to utter a password or a predefined phrase

Abstract

A system includes a user speech profile stored on a computer readable storage device, the speech profile containing a plurality of phonemes with user identifying characteristics for the phonemes, and a speech processor coupled to access the speech profile to generate a phrase containing user distinguishing phonemes based on a difference between the user identifying characteristics for such phonemes and average user identifying characteristics, such that the phrase has discriminability from other users. The speech processor may also or alternatively select the phrase as a function of ambient noise.

Claims

1. A system comprising:

a user speech profile stored on a computer readable storage device, the speech profile containing a plurality of phonemes with user identifying characteristics for the phonemes, wherein the identifying characteristic comprise a representation of the amount of pronunciation difference for each of the plurality of phonemes from an average user; and

a speech processor coupled to access the speech profile and generate a phrase containing user distinguishing phonemes based on the pronunciation difference between the user identifying characteristics for such phonemes and average user identifying characteristics, such that the phrase has discriminability from other users wherein the speech processor generates the phrase by searching for a phrase in a library that contains user distinguishing phonemes, wherein a user distinguishing phoneme is based on pronunciation of a phoneme by a user differing from the way most other users pronounce the phoneme and wherein such distinguishing phonemes have an associated score indicative of a magnitude by which such pronunciation differs and wherein the associated score is adjusted based on a difference of the pronunciation from ambient noise, resulting in a different generated phrase in the presence of the ambient noise.

2. The system of claim 1 and further comprising an audio input to receive the generated phrase spoken by the user.

3. The system of claim 2 and further comprising a display coupled to receive the generated phrase and display it to the user.

4. The system of claim 1 wherein the speech processor further receives the generated phrase spoken by the user via the audio input and verifies whether the phrase was spoken by the user to confirm the identity of the user and allow access to the system.

5. The system of claim 1 wherein the generated phrase contains at least two user distinguishing phonemes.

6. The system of claim 1 wherein the generated phrase contains at least two instances of a user distinguishing phoneme, and wherein the associated score is an average of scores related to ambient noise and user differences in pronunciation.

7. A system comprising:

an audio input to receive speech from a user, the speech including ambient noise proximate the user;

a speech processor coupled to the audio input to generate a phrase containing user distinguishing phonemes as a function of an associated score indicative of a magnitude by which pronunciation of the distinguishing phonemes differs from that of an average user, wherein the associated score includes an ambient noise score such that the phrase has discriminability from the ambient noise, wherein the speech processor generates the phrase by searching for a phrase in a library that contains user distinguishing phonemes, and wherein the associated score is adjusted based on a difference of the pronunciation from ambient noise, resulting in a different generated phase in the presence of the ambient noise.

8. The system of claim 7 wherein the speech processor selects the phrase as a function of a difference between the user identifying characteristics for such phonemes and average user identifying characteristics, such that the phrase has enhanced discriminability from other users.

9. The system of claim 7 and further comprising:

an audio input to receive the generated phrase spoken by the user; and

a display coupled to receive the generated phrase and display it to the user.

10. The system of claim 7 wherein the speech processor further receives the generated phrase spoken by the user via the audio input and verities whether the phrase was spoken by the user to confirm the identity of the user.

11. The system of claim 7 and further comprising a library of phrases stored on a computer readable storage device, and wherein the speech processor generates the phrase by searching for a phrase in the library that contains user distinguishing phonemes.

12. The system of claim 11 wherein the generated phrase contains at least two user distinguishing phonemes.

13. The system of claim 11 wherein the generated phrase contains at least two instances of a user distinguishing phoneme.

14. A method comprising:

receiving a speech audio input from a user, the speech including ambient noise signal proximate the user; and

generating a phrase containing user distinguishing phonemes by a speech processor coupled to the audio input, as a function of an associated score indicative of a magnitude by which pronunciation of the distinguishing phonemes differs from that of an average user, wherein the associated score includes an ambient noise score such that the phrase has discriminability from the ambient noise, wherein the speech processor generates the phrase by searching for a phrase in a library that contains user distinguishing phonemes, and wherein the ambient noise score results in a generation of a different phrase in the presence of the ambient noise.

15. The method of claim 14 wherein the generated phrase is also selected as a function of a difference between the user identifying characteristics for such phonemes and average user identifying characteristics, such that the phrase has enhanced discriminability from other users.

16. The method of claim 14 and further comprising:

displaying the generated phrase to the user;

receiving the generated phrase spoken by the user; and

verifying whether the phrase was spoken by the user to confirm the identity of the user.

17. The method of claim 14 wherein generating the phrase is performed by searching for a phrase in the library that contains user distinguishing phonemes.

18. The method of claim 17 wherein the generated phrase contains at least two user distinguishing phonemes.

19. The system of claim 17 wherein the generated phrase contains at least two instances of a user distinguishing phoneme.