Voice print identification portal

• US 9,799,338 B2
• Filed: 04/28/2014
• Issued: 10/24/2017
• Est. Priority Date: 03/13/2007
• Status: Active Grant

• Current Assignee: Voicelt Technologies LLC
• Original Assignee: Voicelt Technologies LLC
• Inventors: Grover, Noel
• Primary Examiner(s): Desir, Pierre-Louis
• Assistant Examiner(s): Kim, Jonathan
• Application Number:
  US14/263,572
• Publication Number:
  US 20140350932A1
• Time in Patent Office: 1,275 Days
• Field of Search: None
• CPC Class Codes: G06F 21/32 using biometric data, e.g. fingerprints, iris scans or voiceprints
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence
• CPC Class Codes: G10L 17/00 Speaker identification or verification techniques
• CPC Class Codes: G10L 17/06 Decision making techniques; Pattern matching strategies
• CPC Class Codes: H04L 63/0861 using biometrical features, e.g. fingerprint, retina-scan cryptographic mechanisms or cryptographic arrangements for entity authentication using biological data H04L9/3231

Abstract

Systems and methods providing for secure voice print authentication over a network are disclosed herein. During an enrollment stage, a client'"'"'s voice is recorded and characteristics of the recording are used to create and store a voice print. When an enrolled client seeks access to secure information over a network, a sample voice recording is created. The sample voice recording is compared to at least one voice print. If a match is found, the client is authenticated and granted access to secure information.

Systems and methods providing for a dual use voice analysis system are disclosed herein. Speech recognition is achieved by comparing characteristics of words spoken by a speaker to one or more templates of human language words. Speaker identification is achieved by comparing characteristics of a speaker'"'"'s speech to one or more templates, or voice prints. The system is adapted to increase or decrease matching constraints depending on whether speaker identification or speaker recognition is desired.

Claims

1. A method of operating a voice analysis system, comprising:

using a front end interface on a computer system that includes a microphone to initialize a voice analysis system for a user by inputting at least a first voice recording of the user via the microphone to be communicated Over a network connection to a voice analysis computer system that analyzes and stores a template voiceprint for authentication of the user;

using the front end interface to input a second voice recording of the user via the microphone and at least one parameter indicating whether the voice analysis computer system is to operate in a first speaker identification mode or a second word recognition mode to be communicated over the network connection to the voice analysis computer system;

using the voice analysis computer system to analyze the second voice recording,

setting voice analysis constraints to a first level if the parameter indicates the first speaker identification mode, or setting the voice analysis constraints to a word recognition second level if the parameter indicates the second mode;

comparing the second voice recording to the template voiceprint, wherein the comparison is based at least in part on the constraints

if the parameter indicates the first speaker identification mode, communicating over the communication network an indication of authentication of the user to at least one of the front end interface to authenticate the user on the computer system or a back end computer system to authenticate access to the back end computer system if the second voice recording provides a positive match with a voice characteristic of the voice template based on a stricter match requirement; and

if the parameter indicates the second word recognition mode, communicating over the communication network to the front end interface a textual representation of the second voice recording if there is a positive match of words represented in the second voice recording based on a lower match requirement,

wherein using the voice analysis computer system is used to analyze a plurality of second voice recordings and further comprises;

storing a representation of each second voice recording as part of the template voiceprint for the user;

if the parameter indicates the speaker identification first mode, then after there is the positive match and before communicating the indication of authentication of the user, using the voice analysis system to compare the second voice recording with each of the representations of each second voice recording stored as part of the template voiceprint for the user associated with the positive match to determine that the second voice recording has not been used previously for authentication of the user and communicating the indication of authentication of the user only if the second voice recording has not been used previously for authentication of the user.

2. A method of operating a voice analysis system computer system having a memory, a processor and a network connection, comprising:

receiving over the network connection a first voice recording of a user input from a front end interface on a computer system that includes a microphone;

initializing a voice analysis system to authenticate the user by analyzing the first voice recording to create a template voiceprint for authentication associated with the user that includes at least one voice characteristic attributed to the user;

storing the template voiceprint in the memory;

receiving over the network connection a second voice recording of the user input from a front end interface on a computer system that includes a microphone and at least one parameter indicating whether the voice analysis computer system is to operate in a first authentication mode or a second recognition mode;

configuring the voice analysis system in the first authentication mode by setting voice analysis constraints that control the operation of the voice analysis system to a first level, or setting the voice analysis constraints to a second level;

using the voice analysis system to create a representation of the second voice recording,

if the voice analysis system is configured for operation in the first authentication mode, using the voice analysis system to perform a comparison, of the representation of the second voice recording with template voiceprints stored in the memory based at least in part on a stricter match requirement of the voice analysis constraints to determine whether there is a template voiceprint with a positive match and, in response, communicating over the communication network an indication of authentication of the user associated with the template voiceprint with the positive match to at least one of the front end interface from which the second voice recording is received to authenticate the user or to a back end computer system to authenticate access to the back end computer system; and

if the voice analysis system is configured for operation in the second recognition mode, using the voice analysis system to perform a comparison of the representation of the second voice recording with template voiceprints stored in the memory based at least in part on a lower match requirement of the voice analysis constraints to determine whether there is a template voiceprint with a positive match and, in response, using at least one of the voice characteristics of the template voiceprint, with the positive match to perform word recognition of the second voice recording and communicating over the communication network to the front end interface from which the second voice record ng is received a textual representation of the word recognition of the second voice recording,

wherein operating the voice analysis computer system is used to analyze a plurality of second voice recordings and further comprises;

storing each of the voiceprints created based on a second voice recording as part of the template voiceprint for the user associated with the positive match for that voiceprint;

if the parameter indicates the first authentication mode, then after there is the positive match and before communicating the indication of authentication of the user associated with the positive match, using the voice analysis system to further perform a comparison of the voiceprint for the second voice recording with each of the voiceprints created stored as part of the template voiceprints for the user associated with the positive match to determine that the second voice recording has not been used previously for authentication of the user and communicating the indication of authentication of the user only if the second voice recording has not been used previously for identification of the user.

3. A method of operating a voice analysis system having at least a memory, a processor and a network connection with at least one input computer system configured for user input that includes a microphone, comprising:

receiving over the network connection a first voice recording from an input computer system of one or more words spoken by a user;

enrolling, the user in the voice analysis system by analyzing the first voice recording to create and store in the memory a template voiceprint for authentication of the user that includes at least one voice characteristic attributed to the user;

receiving over the network connection a second voice recording from an input computer system and at least one parameter indicating whether the voice analysis computer system is to operate in a first authentication mode or a second recognition mode;

if the voice analysis system is configured for operation in the first authentication mode, using the voice analysis system to compare a representation of the second voice recording with template voiceprints stored in the memory based at least in part on a stricter match requirement of voice analysis constraints of the at least one voice characteristic to determine whether there is a template voiceprint with a positive match and, if so, then;

storing the representation of the second voice recording as part of the template voiceprint for the user associated with the positive match;

comparing the representation of the second voice recording with each part of the template voiceprints for the user associated with the positive match to determine that the second voice recording has not been used previously for authentication of the user; and

communicating over the communication network an indication of authentication of the user associated with the template voiceprint with the positive match only if the second voice recording has not been used previously for authentication of the user; and

if the voice analysis system is configured for operation in the second recognition mode, using the voice analysis system to compare a representation of the second voice recording with template voiceprints stored in the memory based at least in part on a lower match requirement of the voice analysis constraints of the at least one voice characteristic to determine whether there is a template voiceprint with a positive match, and, if so, then;

performing a word recognition of the second voice recording using in part the at least one voice characteristic; and

communicating over the communication network a textual representation of the word recognition of the second voice recording.

4. The method of claim 3 wherein the step of communicating over the communication networks in the first authentication mode includes sending the indication to at least one of the input computer system or a back end computer system.

5. The method of claim 3 wherein the step of communicating over the communication networks in the second recognition mode includes sending the textual representation to at least one of the input computer system or a back end computer system.