System and method for discriminating between speakers for authentication
First Claim
Patent Images
1. A computer-implemented method of controlling ahuman entity'"'"'s access to an account comprising:
- processing biometric information with a speech recognition engine from a plurality of different entities to generate a first set of voice prints for individual ones of such plurality of different entities when articulating utterances of a first spoken text based challenge sentence;
wherein a first voice print is associated with a first entity articulating said first spoken text based challenge sentence for a first user account;
generating a challenge set of additional spoken text based challenge sentences with a computing system for controlling an entity'"'"'s access to the account;
generating a second set of voice prints for said additional spoken text based challenge sentences with a computing system;
wherein said second set of voice prints include both articulated voiceprints and reference voiceprints generated by the computing system for said additional spoken text based challenge sentences;
training said speech recognition engine to distinguish between different entities uttering such challenge set of additional spoken text based challenge sentences;
selecting said first spoken text challenge sentence with the computing system to be articulated as a first speech utterance by an unverified entity accessing the first user account;
storing a first access attempt voice print for the unverified entity at the computing system for the spoken challenge sentence as articulated by the unverified entity;
selecting a first challenge sentence at random from the set of additional spoken text based challenge sentences with the computing system and presenting such sentence to said unverified entity during an access attempt to the first user account;
receiving a second speech utterance from said unverified entity in response to said first challenge sentence;
processing said second speech utterance with said computing system to determine speech characteristics of said unverified entity including a second access attempt voice print; and
processing said first access attempt voice print, said second access attempt voice print, said first voice print, said first set of voice prints and said second set of voice prints with the computing system to determine whether said first speech utterance and second speech utterance originated from said first entity; and
controlling whether said unverified entity is allowed to access the account based on determining whether said unverified entity and said first entity are the same person.
1 Assignment
0 Petitions
Accused Products
Abstract
An audible based electronic challenge system is used to control access to resources by using a test to identify an origin of a voice. The test is based on optimized text sentences selected for their discrimination capability in identifying different speakers.
116 Citations
34 Claims
-
1. A computer-implemented method of controlling a
human entity'"'"'s access to an account comprising: -
processing biometric information with a speech recognition engine from a plurality of different entities to generate a first set of voice prints for individual ones of such plurality of different entities when articulating utterances of a first spoken text based challenge sentence; wherein a first voice print is associated with a first entity articulating said first spoken text based challenge sentence for a first user account; generating a challenge set of additional spoken text based challenge sentences with a computing system for controlling an entity'"'"'s access to the account; generating a second set of voice prints for said additional spoken text based challenge sentences with a computing system;
wherein said second set of voice prints include both articulated voiceprints and reference voiceprints generated by the computing system for said additional spoken text based challenge sentences;training said speech recognition engine to distinguish between different entities uttering such challenge set of additional spoken text based challenge sentences; selecting said first spoken text challenge sentence with the computing system to be articulated as a first speech utterance by an unverified entity accessing the first user account; storing a first access attempt voice print for the unverified entity at the computing system for the spoken challenge sentence as articulated by the unverified entity; selecting a first challenge sentence at random from the set of additional spoken text based challenge sentences with the computing system and presenting such sentence to said unverified entity during an access attempt to the first user account; receiving a second speech utterance from said unverified entity in response to said first challenge sentence; processing said second speech utterance with said computing system to determine speech characteristics of said unverified entity including a second access attempt voice print; and processing said first access attempt voice print, said second access attempt voice print, said first voice print, said first set of voice prints and said second set of voice prints with the computing system to determine whether said first speech utterance and second speech utterance originated from said first entity; and controlling whether said unverified entity is allowed to access the account based on determining whether said unverified entity and said first entity are the same person. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23)
-
-
24. A challenge system for controlling access to an
electronic account on a computing system comprising: -
one or more software routines implemented in a computer readable medium and adapted to cause the challenge system to; process biometric information from a plurality of different entities to generate a first set of voice prints for individual ones of such plurality of different entities when articulating utterances of a first spoken text based challenge sentence; associate a first voice print with a first entity articulating said first spoken text based challenge sentence; generate a challenge set of additional spoken text based challenge sentences for controlling an entity'"'"'s access to the account; generate a second set of voice prints for said additional spoken text based challenge sentences; wherein said second set of voice prints include both articulated voiceprints and reference voiceprints generated for said additional spoken text based challenge sentences; implement a speech recognition engine to distinguish between different entities uttering such challenge set of additional spoken text based challenge sentences; select a first spoken challenge sentence to be articulated as a first speech utterance by an unverified entity accessing the first user account; store a first access attempt voice print for the unverified entity at the computing system for the spoken challenge sentence as articulated by the unverified entity; select a first challenge sentence at random from the set of additional spoken text based challenge sentences and present such sentence to said unverified entity during an access attempt to the first user account; receive a second speech utterance from said unverified entity in response to said first challenge sentence; process said second speech utterance to determine speech characteristics of said unverified entity including a second access attempt voice print; and process said first access attempt voice print, said second access attempt voice print, said first voice print, said first set of voice prints and said second set of voice prints to determine whether said second speech utterance also originated from said first entity; and control whether unverified entity is allowed to access the account based on a determination of whether said unverified entity and said first entity are the same person. - View Dependent Claims (25, 26, 27, 28, 29, 30, 31, 32, 33, 34)
-
Specification