Voice-based verification and identification methods and systems

Claim 3

The method of speaker recognition according to claim 1 wherein said first site is a central location while said second site is a predetermined number of terminal locations.

Claim 4

The method of speaker recognition according to claim 1 further comprising an additional step of: g) transmitting said determination signal from said first site to said second site.

Claim 5

The method of speaker recognition according to claim 1 wherein said step c) further comprising the following steps of: h) sending a selection signal from said first site to said second site; h') identifying a desirable portion and a remaining portion of said voice data in response to said selection signal, said desirable portion substantially reflecting characteristics of said voice data; i) generating said intermediate voice characteristic information based upon said desirable portion; and j) retaining said remaining portion of said voice data.

Claim 6

The method of speaker recognition according to claim 4 wherein said steps i) and j) are repeated after said desirable portion is redefined from said remaining portion.

Claim 7

The method of speaker recognition according to claim 1 wherein said intermediate voice characteristic information is first encoded and then transmitted in said step d) and later decoded in said step e).

Claim 8

The method of speaker recognition according to claim 6 further comprising additional steps of: k) selecting an encryption scheme from a plurality of predetermined encryption schemes; l) transmitting encryption information on said selected encryption scheme from said first site to said second site; m) encoding said intermediate voice characteristic information according to said encryption information; and n) transmitting said encoded intermediate voice characteristic information from said second site to said first site.

Claim 9

The method of speaker recognition according to claim 7 wherein said step e) further comprises an additional step o) of decoding at said first site said encoded intermediate voice characteristic information according to said encryption information.

Claim 10

The method of speaker recognition according to claim 6 wherein said intermediate voice characteristic information is encoded according to a predetermined encoding and decoding scheme.

Claim 11

The method of speaker recognition according to claim 9 wherein one example of said predetermined encoding and decoding scheme includes the following steps of: p) dividing said intermediate voice characteristic information into a predetermined number of sequential segments at said second site; q) transmitting said segments in a predetermined non-sequential order from said second site to said first site; and r) sequencing said transmitted segments of said intermediate voice characteristic information back to said sequential segments at said first site.

Claim 12

The method of speaker recognition according to claim 1 wherein said intermediate voice characteristic information includes a voice data portion containing data indicative of said voice data and a parameter portion containing information on said voice data portion.

Claim 13

The method of speaker recognition according to claim 1 wherein in case said determination signal indicates a failure to match, said step f) further comprising the additional steps of: s) inputting alternative identification information at said second site; t) transmitting said alternative identification to said first site; and u) performing a predetermined alternative verification procedure based upon said alternative identification information.

Claim 14

The method of speaker recognition according to claim 1 wherein said step a) further comprises the following steps prior to storing said first final voice characteristic information: v) inputting test voice data; w) generating test intermediate voice characteristic information based upon said test voice data; and x) further processing said test voice data so as to generate said first final voice characteristic information.

Claim 15

The method of speaker recognition according to claim 1 wherein said step a) further comprises an additional step y) of updating said first voice characteristic information upon confirming that a requester to update said first voice characteristic information has authorization.

Claim 16

The method of speaker recognition according to claim 14 wherein said authorization is confirmed by a predetermined password.

Claim 17

The method of speaker recognition according to claim 15 wherein in case an improper attempt to update said first final voice characteristic information is detected, said improper attempt being recorded.

Claim 18

The method of speaker recognition according to claim 1 wherein said step b) is performed while predetermined noise is played in the background, said step e) removing said predetermined noise.

Claim 19

The method of speaker recognition according to claim 1 further comprising the steps of: I) selecting one noise pattern from a plurality of redetermined number of noise patterns; II) transmitting a noise signal specifying said selected one noise pattern from said first site to said second site prior to said step b); III) inputting said noise pattern and said voice data during said step b) at said second site; and VI) removing said noise pattern at said first site during said step e).

Claim 22

The method of speaker recognition according to claim 19 wherein said first site is a central location while said second site is a predetermined number of terminal locations.

Claim 23

The method of speaker recognition according to claim 20 wherein said determination factor is a threshold value which affects said determination step f), said threshold value being variable.

Claim 24

A speaker recognition system, comprising: a registration unit for processing voice data to generate standard voice characteristic information according said voice data and storing said standard voice characteristic information therein; a first processing unit for inputting test voice data and for processing said test voice data to generate intermediate test voice characteristic information; and a second processing unit communicatively connected to said first processing unit for receiving said intermediate test voice characteristic information and for further processing said intermediate test voice characteristic information to generate test voice characteristic information, said processing unit connected to said registration processing unit for determining if said test voice characteristic information substantially matches said standard voice characteristic information.

Claim 25

A speaker recognition system, comprising: a registration unit for processing voice data to generate standard voice characteristic information according said voice data and storing said standard voice characteristic information therein; a first processing unit for inputting test voice data and for processing said test voice data to generate intermediate test voice characteristic information; and a second processing unit communicatively connected to said first processing unit for receiving said intermediate test voice characteristic information and for further processing said intermediate test voice characteristic information to generate test voice characteristic information, said processing unit connected to said registration processing unit for determining if said test voice characteristic information substantially matches said standard voice characteristic information.

Claim 26

The speaker recognition system according to claim 22 wherein said first processing unit and said second processing unit are a part of computer network, said first processing unit being a terminal while said second processing unit being a central unit.

Claim 27

The speaker recognition system according to claim 22 where said processing unit generates a determination signal indicative of the determination and transmits said determination signal to said first processing unit.

Claim 28

The speaker recognition system according to claim 22 wherein said first processing unit further comprises: a diving unit for dividing said voice data into a desirable portion and a remaining portion, said desirable portion substantially reflecting characteristics of said voice data; an extraction unit for extracting said intermediate voice characteristic information from said desirable portion; and a storage unit for retaining said remaining portion of said voice data.

Claim 29

The speaker recognition system according to claim 22 wherein said first processing unit further includes an encoding unit for encoding said intermediate voice characteristic information and said second processing unit further includes a decoding unit for decoding said encoded intermediate voice characteristic information.

Claim 30

The speaker recognition according to claim 26 wherein said second processing unit selects an encryption scheme from a plurality of predetermined encryption schemes, said second processing unit transmitting encryption information on said selected encryption scheme to said first processing unit, said encoding unit encoding said intermediate voice characteristic information according to said encryption information, said first processing unit transmitting said encoded intermediate voice characteristic information to said second processing unit, said decoding unit decoding said encoded intermediate voice characteristic information according to said encryption information.

Claim 31

The speaker recognition system according to claim 26 wherein said intermediate voice characteristic information is encoded according to a predetermined encoding and decoding scheme.

Claim 32

The speaker recognition system according to claim 22 wherein said intermediate voice characteristic information includes a voice data portion containing data indicative of said voice data and a parameter portion containing information on said voice data portion.

Claim 33

The speaker recognition system according to claim 24 further comprising: an alternative identification input device for inputting alternative identification information in case said determination signal indicates a failure to match, said second processing unit performing a predetermined alternative verification procedure based upon said alternative identification information.

Claim 34

The speaker recognition system according to claim 22 wherein said registration unit further comprises: a password input unit for inputting a current password; a password comparison unit for comparing said current password to a stored password and generating a comparison signal; and an updating unit for updating said first voice characteristic information based upon said comparison signal.

Claim 35

The speaker recognition system according to claim 31 wherein said registration unit records an improper attempt to update said first final voice characteristic information based upon said comparison signal.

Claim 36

The speaker recognition system according to claim 26 wherein said encoding unit generates predetermined noise while inputting test voice data to said first processing unit, said decoding unit removing said predetermined noise.

Claim 37

The speaker recognition system according to claim 26 wherein said second processing unit selects one noise pattern from a plurality of predetermined number of noise patterns, said second processing unit transmitting a noise signal specifying said selected one noise pattern to said encoder, said encoder adding said noise pattern to said voice data, said decoder removing said noise pattern.

Claim 38

A speaker recognition system, comprising: a first processing unit for processing voice data to generate standard voice characteristic information according said voice data and storing said standard voice characteristic information with an associated id information and a determination factor; a second processing unit operationally connected to said first processing unit for inputting said associated id information and test voice data, said second processing unit transmitting to said first processing unit said associated id information, said first processing unit sending said standard voice characteristic information and the determination factor to said second processing unit, said second processing unit generating a test voice characteristic information based upon said test voice data and the determination factor, said second processing unit determining whether or not said standard voice characteristic information substantially matches said test voice characteristic information.

Claim 39

A speaker recognition system, comprising: a first processing unit for processing voice data to generate standard voice characteristic information according said voice data and storing said standard voice characteristic information with an associated id information and a determination factor; a second processing unit operationally connected to said first processing unit for inputting said associated id information and test voice data, said second processing unit transmitting to said first processing unit said associated id information, said first processing unit sending said standard voice characteristic information and the determination factor to said second processing unit, said second processing unit generating a test voice characteristic information based upon said test voice data and the determination factor, said second processing unit determining whether or not said standard voice characteristic information substantially matches said test voice characteristic information.

Claim 40

The speaker recognition system according to claim 35 wherein said first processing unit is a central processor while said second processing unit includes a predetermined number of terminals.

Claim 41

The speaker recognition system according to claim 35 wherein said second processing unit determines if said standard voice characteristic information substantially matches said test voice characteristic information based upon a threshold value.

Claim 42

The speaker recognition system according to claim 37 wherein said first processing unit sends a threshold value to said second processing unit.