CA2267954C - Speaker verification method - Google Patents
Speaker verification method Download PDFInfo
- Publication number
- CA2267954C CA2267954C CA002267954A CA2267954A CA2267954C CA 2267954 C CA2267954 C CA 2267954C CA 002267954 A CA002267954 A CA 002267954A CA 2267954 A CA2267954 A CA 2267954A CA 2267954 C CA2267954 C CA 2267954C
- Authority
- CA
- Canada
- Prior art keywords
- user
- subscriber
- code
- voice
- entered
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04M—TELEPHONIC COMMUNICATION
- H04M3/00—Automatic or semi-automatic exchanges
- H04M3/42—Systems providing special services or facilities to subscribers
- H04M3/487—Arrangements for providing information services, e.g. recorded voice services or time announcements
- H04M3/493—Interactive information services, e.g. directory enquiries ; Arrangements therefor, e.g. interactive voice response [IVR] systems or voice portals
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07C—TIME OR ATTENDANCE REGISTERS; REGISTERING OR INDICATING THE WORKING OF MACHINES; GENERATING RANDOM NUMBERS; VOTING OR LOTTERY APPARATUS; ARRANGEMENTS, SYSTEMS OR APPARATUS FOR CHECKING NOT PROVIDED FOR ELSEWHERE
- G07C9/00—Individual registration on entry or exit
- G07C9/30—Individual registration on entry or exit not involving the use of a pass
- G07C9/32—Individual registration on entry or exit not involving the use of a pass in combination with an identity check
- G07C9/37—Individual registration on entry or exit not involving the use of a pass in combination with an identity check using biometric data, e.g. fingerprints, iris scans or voice recognition
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L17/00—Speaker identification or verification
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04M—TELEPHONIC COMMUNICATION
- H04M3/00—Automatic or semi-automatic exchanges
- H04M3/38—Graded-service arrangements, i.e. some subscribers prevented from establishing certain connections
- H04M3/382—Graded-service arrangements, i.e. some subscribers prevented from establishing certain connections using authorisation codes or passwords
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04M—TELEPHONIC COMMUNICATION
- H04M2201/00—Electronic components, circuits, software, systems or apparatus used in telephone systems
- H04M2201/40—Electronic components, circuits, software, systems or apparatus used in telephone systems using speech recognition
Abstract
Method for enabling a user of a system to obtain access to one or more services, for example for enabling a caller to access some information on a remote server. The method comprises the following steps: the system first prompts the caller to identify himself by entering a subscriber code (121). A speaker verification is then made using a voice comparison between the caller's voice and a voice reference (123) associated with the subscriber the caller claims to be. Access to the requested services is given if the outcome of said voice comparison is positive; if the system does not store a sufficiently good voice reference to verify the identity of the subscriber the user claim to be with a predetermined degree of reliability, the caller is prompted to enter a confidential PIN code (122). If the PIN code entered is correct, access to the requested services is given and a new voice reference is stored.
Description
Speaker Verification Method Field of the Invention The invention relates to a method for enabling a user of a system to obtain access to one or more services.
Back4round of the Invention In many situations, it is necessary to identify users before giving them access to services or systems. These services include remote access to computers, electronic banking transactions, security access control for buildings, etc. In the same way, for many new telephone services it is necessary to verify unambiguously the identity of the caller. The caller identity usually cannot be determined from the access point in the telephone network used by the caller, as a caller may want to access those services from different access points.
An object of the present invention is to provide a method for enabling a user of a system to be identified and thus to obtain access to one or more services in a convenient way.
In a well-known method used for identifying a user in a system, the user is first requested to identify himself by entering a personal identification code (ID). The user is then requested to enter a secret password used to verify the claimed identity. Access to the requested services is given only if the entered password matches the entered ID. This is the method, for example, for using the calling cards proposed by many different telephone network operators. A
subscriber who wants to use the service is usually furnished with a card, for example in credit card format, on which a subscriber code or number is printed.
Furthermore, the subscriber is given a confidential personal identification code (PIN code) he has to remember. The subscriber code is needed for identifying the subscriber and the PIN code for verifying the entered identity.
To use the service, the subscriber calls a toll-free telephone number which connects him to the server which implements the calling card service.
Once the call setup has been established, a speech generator in the server prompts the caller to enter first his subscriber number (which is printed on the card he has previously bought), and then his confidential PIN code. Those numbers may be entered manually on the telephone keyboard or, more com-fortably, orally. In this latter case, the digits are recognized by a speech recog-nizing algorithm implemented in the server.
If the server recognizes the subscriber number entered and if the s entered PIN code matches this subscriber number, the subscriber can then be connected to another subscriber within the network. The invoice for the tele-phone call will be sent automatically to the calling subscriber, independently of the place from which the calf was made.
This method is very tedious for the subscriber who has to enter suc-io cessively two numbers, one of which has to be memorized. The security is not optimal, as very often the PIN code is written down near or on the card.
The U.S. Patent No.5,297,194 describes a method in which the sub-scriber is prompted to enter one unique password digit-by-digit. A speech rec-ognition algorithm running on the digital processing system of the server ~s processes the information received and, using a statistical recognition strategy, determines the digits. After all digits have been recognized, and if the password is valid, the server uses a speaker verification algorithm to determine if the entered password has been spoken by a caller whose voice has been previ-ously assigned to the password. Access to other services is allowed only if the 2o verification algorithm establishes a match; otherwise, the system rejects the access inquiry and the call is terminated.
This method generally requires the subscribers to have previously enrolled in the system. Accordingly, the speaker verification algorithm uses a voice reference database comprising voice references collected from all 2s authorized subscribers. Enrollment is accomplished by having the subscriber repeat a multidigit password several times. A voice reference is then obtained from those utterances. This voice reference is assigned to the password attrib-uted to the subscriber.
This method requires active collaboration of the user at the time of 3o enrollment and thus discourages in particular new subscribers. Moreover en-rollment is generally carried out in a single session from a single telephone apparatus, for example from the telephone at the address of the new subscriber'.
The quality of the voice reference registered therefore very much depends upon the characteristics of the telephone apparatus originally used, and the subscriber risks not being recognized if he later calls from a different telephone.
Other speaker verification method necessitating an enrollment session are described in USA. 4,827,518 and in the AT&T Technical Journal, Vol. 65, No~.
5, September 1986, pages 68-74, Birnbaum Martha: "A voice password system for access security".
Other speaker recognition methods and algorithms are described by George R. Doddington in "Speaker Recognition--Identifying People by their Voices", Proceedings of the IEEE, vol. 73, No. 11, November 1985, pp.
1651-1664; by John Oglesby in "What's in a number? Moving beyond the equal error rate", Speech Communication, vol. 17, 1995, pp. 193-208; and by Sadaoki I=urui, in "An overview of Speaker Recognition Technology", ESCA Workshop on Automatic Speaker Recognition, Identification and Verification.
Summary of the Invention The object of the present invention is to provide an improved method for enabling a user of a system to obtain access to one or more services.
According to the invention, a subscriber receives a subscriber code and a confidential PIN code, as in the prior art. The subscriber code serves to identify the subscriber and the PIN code to verify the claimed identity.
To use the service, the user is first prompted (for example by a speech generator) to enter his subscriber code. This number may, for example, be printed on a subscriber card he has previously bought. A speaker verification algorithm is then used for verifying the identity claimed by the user by making a voice comparison between the user's voice and a voice reference stored in the system and associated with the subscriber code entered by the user. Access to the requested services is given if the user's voice matches said voice reference.
Back4round of the Invention In many situations, it is necessary to identify users before giving them access to services or systems. These services include remote access to computers, electronic banking transactions, security access control for buildings, etc. In the same way, for many new telephone services it is necessary to verify unambiguously the identity of the caller. The caller identity usually cannot be determined from the access point in the telephone network used by the caller, as a caller may want to access those services from different access points.
An object of the present invention is to provide a method for enabling a user of a system to be identified and thus to obtain access to one or more services in a convenient way.
In a well-known method used for identifying a user in a system, the user is first requested to identify himself by entering a personal identification code (ID). The user is then requested to enter a secret password used to verify the claimed identity. Access to the requested services is given only if the entered password matches the entered ID. This is the method, for example, for using the calling cards proposed by many different telephone network operators. A
subscriber who wants to use the service is usually furnished with a card, for example in credit card format, on which a subscriber code or number is printed.
Furthermore, the subscriber is given a confidential personal identification code (PIN code) he has to remember. The subscriber code is needed for identifying the subscriber and the PIN code for verifying the entered identity.
To use the service, the subscriber calls a toll-free telephone number which connects him to the server which implements the calling card service.
Once the call setup has been established, a speech generator in the server prompts the caller to enter first his subscriber number (which is printed on the card he has previously bought), and then his confidential PIN code. Those numbers may be entered manually on the telephone keyboard or, more com-fortably, orally. In this latter case, the digits are recognized by a speech recog-nizing algorithm implemented in the server.
If the server recognizes the subscriber number entered and if the s entered PIN code matches this subscriber number, the subscriber can then be connected to another subscriber within the network. The invoice for the tele-phone call will be sent automatically to the calling subscriber, independently of the place from which the calf was made.
This method is very tedious for the subscriber who has to enter suc-io cessively two numbers, one of which has to be memorized. The security is not optimal, as very often the PIN code is written down near or on the card.
The U.S. Patent No.5,297,194 describes a method in which the sub-scriber is prompted to enter one unique password digit-by-digit. A speech rec-ognition algorithm running on the digital processing system of the server ~s processes the information received and, using a statistical recognition strategy, determines the digits. After all digits have been recognized, and if the password is valid, the server uses a speaker verification algorithm to determine if the entered password has been spoken by a caller whose voice has been previ-ously assigned to the password. Access to other services is allowed only if the 2o verification algorithm establishes a match; otherwise, the system rejects the access inquiry and the call is terminated.
This method generally requires the subscribers to have previously enrolled in the system. Accordingly, the speaker verification algorithm uses a voice reference database comprising voice references collected from all 2s authorized subscribers. Enrollment is accomplished by having the subscriber repeat a multidigit password several times. A voice reference is then obtained from those utterances. This voice reference is assigned to the password attrib-uted to the subscriber.
This method requires active collaboration of the user at the time of 3o enrollment and thus discourages in particular new subscribers. Moreover en-rollment is generally carried out in a single session from a single telephone apparatus, for example from the telephone at the address of the new subscriber'.
The quality of the voice reference registered therefore very much depends upon the characteristics of the telephone apparatus originally used, and the subscriber risks not being recognized if he later calls from a different telephone.
Other speaker verification method necessitating an enrollment session are described in USA. 4,827,518 and in the AT&T Technical Journal, Vol. 65, No~.
5, September 1986, pages 68-74, Birnbaum Martha: "A voice password system for access security".
Other speaker recognition methods and algorithms are described by George R. Doddington in "Speaker Recognition--Identifying People by their Voices", Proceedings of the IEEE, vol. 73, No. 11, November 1985, pp.
1651-1664; by John Oglesby in "What's in a number? Moving beyond the equal error rate", Speech Communication, vol. 17, 1995, pp. 193-208; and by Sadaoki I=urui, in "An overview of Speaker Recognition Technology", ESCA Workshop on Automatic Speaker Recognition, Identification and Verification.
Summary of the Invention The object of the present invention is to provide an improved method for enabling a user of a system to obtain access to one or more services.
According to the invention, a subscriber receives a subscriber code and a confidential PIN code, as in the prior art. The subscriber code serves to identify the subscriber and the PIN code to verify the claimed identity.
To use the service, the user is first prompted (for example by a speech generator) to enter his subscriber code. This number may, for example, be printed on a subscriber card he has previously bought. A speaker verification algorithm is then used for verifying the identity claimed by the user by making a voice comparison between the user's voice and a voice reference stored in the system and associated with the subscriber code entered by the user. Access to the requested services is given if the user's voice matches said voice reference.
If said voice reference is not good enough to verify the claimed identity with a prescribed degree of reliability, the user is then prompted to enter his confidential PIN code. The PIN code is then checked, and, if the outcome of this check is positive, a new voice reference is created in a background process and associated with the subscriber code entered by the user.
One advantage of the invention, especially if used for identifying callers in a telephone network, is that the voice reference is created and stored during several sessions, usually from different locations and with various telephone apparatus. Thus, the voice reference stored for a given subscriber usually takes very different environmental conditions (noise, telephone lines, microphone, etc.) into account.
According to the invention, a specific enrollment session is not needed.
Therefore, the method is very user-friendly, especially for new sub-scribers.
In the same way, the voice reference created takes into account possible user speech variations between two sessions. For example, it happens often that the user's voice is less sure and quick during the first session than during subsequent ones, when the user feels more confident.
It is possible to use a very high number of utterances for the voice reference without making the procedure too uncomfortable for the user. Even if the voice reference is already sufficiently good for a reliable verification of the user's identity, it is still possible to take new utterances into account in order to improve even more the quality of the voice reference and to adapt it to the user's voice variations.
According to one aspect of the invention, there is provided a method for enabling a user of a system to obtain access to one or more services that comprises prompting the user to identify himself with a subscriber code.
Provided that the subscriber code is registered and a voice reference has been assigned to the subscriber code, the subscriber's identity is verified by comparing the user's voice to the voice reference associated with the subscriber code entered by the user. Access to the requested services is permitted provided the user's voice matches the voice reference. When no voice reference is yet associated with the subscriber code entered by the user, or when the voice reference associated with the subscriber code entered by the user is not yet sufficiently good to verify the subscriber's identity with a predetermined degree of reliability, the user is prompted to enter a PIN code.
The PIN code entered by the user is then compared with a PIN code associated 5 with the subscriber code entered by the user. If the PIN code entered by the user matches the PIN code associated with the subscriber code entered by the user, access to the requested services is gained. Where no voice reference was previously on file, a new voice reference is created and associated with the subscriber code, and where a voice reference was previously on file, an improved voice reference is created and associated with the subscriber code.
Brief Description of the Drawings An embodiment of the invention is described below, by way of example only, and illustrated by the figures which show:
Figure 1 is a flow chart of this embodiment, Figure 2 is a partial flow chart of the steps undertaken by the server when it has not stored sufficient reference voice parameters to recognize the subscriber with sufficient security, Figure 3 is a partial flow chart of the steps undertaken by the server when it has already stored sufficiently good reference voice parameters to recognize the subscriber with a prescribed degree of reliability, Figure 4 shows the structure of a database set up according to the method of the present invention.
Detailed Description of the Preferred Embodiment Before the different steps of the method according to the invention are explained in detail, a database 120, used for implementing the method according to the inventian, will be described, with reference to Figure 4.
Each record of the subscriber database corresponds to one subscriber and contains all information needed to identify and verify the identity of the user. The associated record for each subscriber includes a subscriber code field 121, a PIN is code field 122, a voice reference field 123 and a reference flag 124.
The 5a subscriber code field 121 includes enough digits to designate unambiguously each subscriber. In this example, the subscriber code field includes alphanumerical digits; however, in many applications, only numbers will be allowed. PIN code field 122 includes enough digits to verify the claimed identity of the user with a prescribed degree of reliability which may depend on the application. Again, digits can be either alphanumerical or just numerical. The voice reference field 123 stores a voice reference corresponding to the subscriber. Depending on the speaker verification algorithm used, the voice reference field may store several samples of the voice (utterances) of the subscriber, one sample extracted from one or several utterances, or one or several parameters extracted from one or several previous utterances of the subscriber. Reference flag 124 is a single bit indicating if the voice reference stored in field 123 is of sufficiently good quality to verify the identity of the subscriber with the prescribed degree of reliability.
One advantage of the invention, especially if used for identifying callers in a telephone network, is that the voice reference is created and stored during several sessions, usually from different locations and with various telephone apparatus. Thus, the voice reference stored for a given subscriber usually takes very different environmental conditions (noise, telephone lines, microphone, etc.) into account.
According to the invention, a specific enrollment session is not needed.
Therefore, the method is very user-friendly, especially for new sub-scribers.
In the same way, the voice reference created takes into account possible user speech variations between two sessions. For example, it happens often that the user's voice is less sure and quick during the first session than during subsequent ones, when the user feels more confident.
It is possible to use a very high number of utterances for the voice reference without making the procedure too uncomfortable for the user. Even if the voice reference is already sufficiently good for a reliable verification of the user's identity, it is still possible to take new utterances into account in order to improve even more the quality of the voice reference and to adapt it to the user's voice variations.
According to one aspect of the invention, there is provided a method for enabling a user of a system to obtain access to one or more services that comprises prompting the user to identify himself with a subscriber code.
Provided that the subscriber code is registered and a voice reference has been assigned to the subscriber code, the subscriber's identity is verified by comparing the user's voice to the voice reference associated with the subscriber code entered by the user. Access to the requested services is permitted provided the user's voice matches the voice reference. When no voice reference is yet associated with the subscriber code entered by the user, or when the voice reference associated with the subscriber code entered by the user is not yet sufficiently good to verify the subscriber's identity with a predetermined degree of reliability, the user is prompted to enter a PIN code.
The PIN code entered by the user is then compared with a PIN code associated 5 with the subscriber code entered by the user. If the PIN code entered by the user matches the PIN code associated with the subscriber code entered by the user, access to the requested services is gained. Where no voice reference was previously on file, a new voice reference is created and associated with the subscriber code, and where a voice reference was previously on file, an improved voice reference is created and associated with the subscriber code.
Brief Description of the Drawings An embodiment of the invention is described below, by way of example only, and illustrated by the figures which show:
Figure 1 is a flow chart of this embodiment, Figure 2 is a partial flow chart of the steps undertaken by the server when it has not stored sufficient reference voice parameters to recognize the subscriber with sufficient security, Figure 3 is a partial flow chart of the steps undertaken by the server when it has already stored sufficiently good reference voice parameters to recognize the subscriber with a prescribed degree of reliability, Figure 4 shows the structure of a database set up according to the method of the present invention.
Detailed Description of the Preferred Embodiment Before the different steps of the method according to the invention are explained in detail, a database 120, used for implementing the method according to the inventian, will be described, with reference to Figure 4.
Each record of the subscriber database corresponds to one subscriber and contains all information needed to identify and verify the identity of the user. The associated record for each subscriber includes a subscriber code field 121, a PIN is code field 122, a voice reference field 123 and a reference flag 124.
The 5a subscriber code field 121 includes enough digits to designate unambiguously each subscriber. In this example, the subscriber code field includes alphanumerical digits; however, in many applications, only numbers will be allowed. PIN code field 122 includes enough digits to verify the claimed identity of the user with a prescribed degree of reliability which may depend on the application. Again, digits can be either alphanumerical or just numerical. The voice reference field 123 stores a voice reference corresponding to the subscriber. Depending on the speaker verification algorithm used, the voice reference field may store several samples of the voice (utterances) of the subscriber, one sample extracted from one or several utterances, or one or several parameters extracted from one or several previous utterances of the subscriber. Reference flag 124 is a single bit indicating if the voice reference stored in field 123 is of sufficiently good quality to verify the identity of the subscriber with the prescribed degree of reliability.
The method used by the system will now be explained with reference to Figures 1 to 3, which relate more particularly to caller identification in a tele-phone network.
A user wishing access to the services offered calls the server imple-s menting the method of the present invention (step 2). During step 4, the user is first greeted and then prompted to enter his subscriber code (or some other form of identification data).
Preferably, the subscriber code is entered orally, for example se-quentially and digit-by-digit. Alternatively, the subscriber code may be entered io manually with a keyboard, for example on the user's receiver. In this latter case, the user will be prompted to say something, for example to repeat a standard sentence.
The next step 6 consists in subscriber identification. If the subscriber code is entered orally, subscriber identification implies speech recognition to is extract the subscriber code from the sentence spoken by the user. If the sub-scriber code is entered manually using the keyboard on the receiver, the gen-erated pulses or tones are used to determine the digits of the subscriber code.
Steps 4 and 6 could possibly be repeated several times until the transmission quality is sufficiently good for identifying the subscriber code. As a different 2o subscriber code is assigned to each different subscriber, after step 6 the sys-tem knows the identity claimed by the user.
A test is made during step 8 to check if a voice reference 123 has al-ready been assigned in database 120 to the subscriber the user claims to be.
If the outcome of this test is negative, phase A is executed, described below in 2s connection with Figure 2. On the other hand, if a voice reference 123 has al-ready been assigned to the identified subscriber, the reference flag 124 is checked during step 10. We will see below that this flag is set to 1 if, and only if, the voice reference 123 assigned to the subscriber is sufficiently good (depending on the algorithm used) in order to verify his identity with the desired 3o degree of reliability.
WO 98/16906 PCTlEP96/04480 If the reference flag 124 was at 0, phase A, discussed below in rela-tion to figure 2, is executed (step 20); on the other hand, if it was set to a posi-tive value, phase B, discussed below in relation to figure 3, is executed (step 30).
s The reference flag 124 can also be set to 0 each time a new sub-scriber is stored in the database. In this case, it is possible to bypass the test 8.
Phase A will now be described in relation to figure 2. Phase A is executed when no voice reference 123 has been assigned to the subscriber in the database 120, or when the voice reference stored is not sufficiently good to to verify the identity of the subscriber with the prescribed degree of reliability. After a few calls, the identity of the subscriber will be verified automatically from the subscriber's voice, and this phase will no longer be executed.
At beginning of phase A, during step 21, the user is prompted to enter his PIN code. This PIN code is confidential; it can either be entered orally is or preferably manually using the keyboard on the receiver. In the first case, a speech recognition algorithm is used to recognize the orally entered PIN code;
in the second case, the PIN code is determined from selection pulses or tones generated manually with the keyboard. During step 22, a verification is made to check if the PIN code entered by the user matches the PIN code 122 assigned 2o to the subscriber the user claims to be. If the PIN code is wrong, access to services is denied and the call is terminated (step 23). In a variant embodiment, the user may have a second or even a third chance to re-enter his PIN code if the first try failed.
On the other hand, if the PIN code is correct, access is given to the 2s requested service (step 25). Depending on the application, the service can be connection with another network subscriber, access to a database, electronic funds transfer, credit card validation, subscriber personal identification, invoic-ing of the call on the user's account with the network provider, etc. However, it should be noted that the invention is not limited to these applications, which are 3o given as examples only.
__._.~.. ...... _ _._. ._. ._.~..~a.~...a m~~_...~______~__._._.__.~..,_..-,.~._._ .. ~._._.~...._.~.~__..
A user wishing access to the services offered calls the server imple-s menting the method of the present invention (step 2). During step 4, the user is first greeted and then prompted to enter his subscriber code (or some other form of identification data).
Preferably, the subscriber code is entered orally, for example se-quentially and digit-by-digit. Alternatively, the subscriber code may be entered io manually with a keyboard, for example on the user's receiver. In this latter case, the user will be prompted to say something, for example to repeat a standard sentence.
The next step 6 consists in subscriber identification. If the subscriber code is entered orally, subscriber identification implies speech recognition to is extract the subscriber code from the sentence spoken by the user. If the sub-scriber code is entered manually using the keyboard on the receiver, the gen-erated pulses or tones are used to determine the digits of the subscriber code.
Steps 4 and 6 could possibly be repeated several times until the transmission quality is sufficiently good for identifying the subscriber code. As a different 2o subscriber code is assigned to each different subscriber, after step 6 the sys-tem knows the identity claimed by the user.
A test is made during step 8 to check if a voice reference 123 has al-ready been assigned in database 120 to the subscriber the user claims to be.
If the outcome of this test is negative, phase A is executed, described below in 2s connection with Figure 2. On the other hand, if a voice reference 123 has al-ready been assigned to the identified subscriber, the reference flag 124 is checked during step 10. We will see below that this flag is set to 1 if, and only if, the voice reference 123 assigned to the subscriber is sufficiently good (depending on the algorithm used) in order to verify his identity with the desired 3o degree of reliability.
WO 98/16906 PCTlEP96/04480 If the reference flag 124 was at 0, phase A, discussed below in rela-tion to figure 2, is executed (step 20); on the other hand, if it was set to a posi-tive value, phase B, discussed below in relation to figure 3, is executed (step 30).
s The reference flag 124 can also be set to 0 each time a new sub-scriber is stored in the database. In this case, it is possible to bypass the test 8.
Phase A will now be described in relation to figure 2. Phase A is executed when no voice reference 123 has been assigned to the subscriber in the database 120, or when the voice reference stored is not sufficiently good to to verify the identity of the subscriber with the prescribed degree of reliability. After a few calls, the identity of the subscriber will be verified automatically from the subscriber's voice, and this phase will no longer be executed.
At beginning of phase A, during step 21, the user is prompted to enter his PIN code. This PIN code is confidential; it can either be entered orally is or preferably manually using the keyboard on the receiver. In the first case, a speech recognition algorithm is used to recognize the orally entered PIN code;
in the second case, the PIN code is determined from selection pulses or tones generated manually with the keyboard. During step 22, a verification is made to check if the PIN code entered by the user matches the PIN code 122 assigned 2o to the subscriber the user claims to be. If the PIN code is wrong, access to services is denied and the call is terminated (step 23). In a variant embodiment, the user may have a second or even a third chance to re-enter his PIN code if the first try failed.
On the other hand, if the PIN code is correct, access is given to the 2s requested service (step 25). Depending on the application, the service can be connection with another network subscriber, access to a database, electronic funds transfer, credit card validation, subscriber personal identification, invoic-ing of the call on the user's account with the network provider, etc. However, it should be noted that the invention is not limited to these applications, which are 3o given as examples only.
__._.~.. ...... _ _._. ._. ._.~..~a.~...a m~~_...~______~__._._.__.~..,_..-,.~._._ .. ~._._.~...._.~.~__..
In a background process (step 24), and simultaneously with step 25, a new voice reference is created using the user's voice sample (utterance) and assigned to the subscriber (field 123). Step 26 is a test to check if the voice ref-erence created is now sufficiently good to verify the subscriber with the s prescribed degree of reliability. What is meant by "sufficiently good"
depends on the algorithm used for voice verification; the prescribed degree of reliability obviously depends on the application. It is possible for instance to determine that the voice reference is sufficiently good if it has been created from a minimum number of utterances, for example ten. In this case, the voice refer-~o ence associated with each user is regarded as good enough as soon as said user has had access to the system 10 times or more. Counting means are nec-essary which store in an additional field associated with each subscriber code 121 the number of accesses entered by this subscriber. Alternatively, it is pos-sible according to the speaker verification algorithm used to determine directly is from certain parameters of the voice reference 123 whether the latter is suffi-ciently good. In the case of a negative response, the reference flag 124 corresponding to the subscriber is set or kept at zero in the course of step 27.
On the other hand, if the voice reference is now sufficient, the reference flag 124 is set to 1 in the course of step 28. In this case, at the time of subsequent 2o calls of the subscriber, only the subscriber code is required, as a rule, and, in-stead of phase A, the system carries out phase B, which will now be discussed in relation to Figure 3.
Phase B is carried out for calls of subscribers for whom the voice is sufficiently known to permit a sufficiently reliable verification of identity.
In the 2s course of step 31, the subscriber identity claimed by the user is verified using a speaker verification algorithm. In so doing, a comparison is made between the user's voice utterance and the voice reference 123 assigned in the system to the claimed subscriber. Any algorithm for verification of a known person can be used, for example an algorithm such as is described in one of the aforemen-3o tinned documents to verify whether the user's utterance matches the voice ref-erence associated with the subscriber code 121 entered by the user.
If the outcome of this test (step 32) is alright, i.e. if the identity of the user has been verified, access is given to the requested services {step 35).
If, on the other hand, the entered voice cannot be matched substantially to the voice reference assigned to the claimed subscriber, access to the requested services is denied and the call is terminated (step 33). In a variant embodiment, phase A is executed, instead of step 33, when the user's identity cannot be s verified from his voice. In this case, as described above, the user is prompted to enter his PIN code.
In a preferred embodiment of the invention, the user's utterance is used to create a new voice reference 123 in a background step 34, even if the voice reference assigned to the subscriber is already sufficiently good. This to new voice reference is stored in field 123 associated with the subscriber code 121 entered by the user. In this way the system adapts progressively to varia-tions in the voice of the subscriber owing to aging, seasonal temperature varia-tions, different telephone apparatus, etc.
The above description relates in particular to use of the method in a is telephone network. It is clear, however, that the invention can also be applied to other types of transactions or networks. For example, the invention can also be used for the verification of identity in automatic teller machines or in access control devices.
._.r _...._. __ ...... _. ____. . __ _.._ _._. _. __.r. _ ...__~-...~..
depends on the algorithm used for voice verification; the prescribed degree of reliability obviously depends on the application. It is possible for instance to determine that the voice reference is sufficiently good if it has been created from a minimum number of utterances, for example ten. In this case, the voice refer-~o ence associated with each user is regarded as good enough as soon as said user has had access to the system 10 times or more. Counting means are nec-essary which store in an additional field associated with each subscriber code 121 the number of accesses entered by this subscriber. Alternatively, it is pos-sible according to the speaker verification algorithm used to determine directly is from certain parameters of the voice reference 123 whether the latter is suffi-ciently good. In the case of a negative response, the reference flag 124 corresponding to the subscriber is set or kept at zero in the course of step 27.
On the other hand, if the voice reference is now sufficient, the reference flag 124 is set to 1 in the course of step 28. In this case, at the time of subsequent 2o calls of the subscriber, only the subscriber code is required, as a rule, and, in-stead of phase A, the system carries out phase B, which will now be discussed in relation to Figure 3.
Phase B is carried out for calls of subscribers for whom the voice is sufficiently known to permit a sufficiently reliable verification of identity.
In the 2s course of step 31, the subscriber identity claimed by the user is verified using a speaker verification algorithm. In so doing, a comparison is made between the user's voice utterance and the voice reference 123 assigned in the system to the claimed subscriber. Any algorithm for verification of a known person can be used, for example an algorithm such as is described in one of the aforemen-3o tinned documents to verify whether the user's utterance matches the voice ref-erence associated with the subscriber code 121 entered by the user.
If the outcome of this test (step 32) is alright, i.e. if the identity of the user has been verified, access is given to the requested services {step 35).
If, on the other hand, the entered voice cannot be matched substantially to the voice reference assigned to the claimed subscriber, access to the requested services is denied and the call is terminated (step 33). In a variant embodiment, phase A is executed, instead of step 33, when the user's identity cannot be s verified from his voice. In this case, as described above, the user is prompted to enter his PIN code.
In a preferred embodiment of the invention, the user's utterance is used to create a new voice reference 123 in a background step 34, even if the voice reference assigned to the subscriber is already sufficiently good. This to new voice reference is stored in field 123 associated with the subscriber code 121 entered by the user. In this way the system adapts progressively to varia-tions in the voice of the subscriber owing to aging, seasonal temperature varia-tions, different telephone apparatus, etc.
The above description relates in particular to use of the method in a is telephone network. It is clear, however, that the invention can also be applied to other types of transactions or networks. For example, the invention can also be used for the verification of identity in automatic teller machines or in access control devices.
._.r _...._. __ ...... _. ____. . __ _.._ _._. _. __.r. _ ...__~-...~..
Claims (7)
1. Method for enabling a user of a system to obtain access to one or more services, comprising:
prompting (4, 6) the user to identify himself with a subscriber code (121);
provided that said subscriber code (121) is registered and a voice reference (123) has been assigned to said subscriber code (121), verifying the subscriber's identity by comparing the user's voice to said voice reference (123) associated with the subscriber code (121) entered by the user, access to the requested services being given if the user's voice matches said voice reference (123);
wherein the following are undertaken during an initial phase when no voice references are yet associated with the subscriber code (121) entered by the user, or when the voice reference (123) associated with the subscriber code entered by the user is not yet sufficiently good to verify the subscriber's identify with a predetermined degree of reliability;
prompting (21) said user to enter PIN code (122);
comparing (22) said PIN code entered by the user with a PIN code associated with the subscriber code (121) entered by the user;
if the PIN code entered by the user matches the PIN code associated with the subscriber code (121) entered by the user, giving the user access to the requested services, creating a new voice reference (123) and associating said new voice reference (123) with said subscriber code (121), or, provided that a voice reference (123) is already associated with the said subscriber code (121), improving said voice reference (123) by said new voice reference.
prompting (4, 6) the user to identify himself with a subscriber code (121);
provided that said subscriber code (121) is registered and a voice reference (123) has been assigned to said subscriber code (121), verifying the subscriber's identity by comparing the user's voice to said voice reference (123) associated with the subscriber code (121) entered by the user, access to the requested services being given if the user's voice matches said voice reference (123);
wherein the following are undertaken during an initial phase when no voice references are yet associated with the subscriber code (121) entered by the user, or when the voice reference (123) associated with the subscriber code entered by the user is not yet sufficiently good to verify the subscriber's identify with a predetermined degree of reliability;
prompting (21) said user to enter PIN code (122);
comparing (22) said PIN code entered by the user with a PIN code associated with the subscriber code (121) entered by the user;
if the PIN code entered by the user matches the PIN code associated with the subscriber code (121) entered by the user, giving the user access to the requested services, creating a new voice reference (123) and associating said new voice reference (123) with said subscriber code (121), or, provided that a voice reference (123) is already associated with the said subscriber code (121), improving said voice reference (123) by said new voice reference.
2. Method according to claim 1, in which the user is prompted to enter his subscriber code (121) orally, a speech recognition algorithm being used to recognize the subscriber code entered.
3. Method according to claim 1, in which the user is prompted to enter his subscriber code (121) on a keyboard, tones or pulses generated by typing the keyboard being used to recognize the subscriber code, and in which the user is prompted to speak a sentence allowing speaker verification.
4. Method according to any one of claims 1 to 3, in which the user is prompted to enter his PIN code (122) when the user's voice does not match said voice reference (123) associated with the subscriber code (121) entered by the user, access to the requested services being given if the PIN code entered by the user matches the PIN code (122) associated with the subscriber code (121) entered by the user.
5. Method according to any one of claims 1 to 4, in which the user is prompted to enter his PIN code (122) orally, a speech recognition algorithm being used to recognize the PIN code entered.
6. Method according to any one of claims 1 to 5, in which the user is prompted to enter his PIN code (122) on a keyboard, tones or pulses generated by typing the keyboard being used to recognize the subscriber code entered.
7. Method according to any one of claims 1 to 6, in which a new voice reference (123) is created (34) and associated with the subscriber code (121) entered by the user when the user's voice matches said voice reference (123) associated with the subscriber code (121) entered by the user.
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/EP1996/004480 WO1998016906A1 (en) | 1996-10-15 | 1996-10-15 | Speaker verification method |
Publications (2)
Publication Number | Publication Date |
---|---|
CA2267954A1 CA2267954A1 (en) | 1998-04-23 |
CA2267954C true CA2267954C (en) | 2005-06-28 |
Family
ID=8166367
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CA002267954A Expired - Fee Related CA2267954C (en) | 1996-10-15 | 1996-10-15 | Speaker verification method |
Country Status (9)
Country | Link |
---|---|
US (1) | US6556127B1 (en) |
EP (1) | EP0932885B1 (en) |
JP (1) | JP3904608B2 (en) |
AT (1) | ATE227868T1 (en) |
AU (1) | AU7293096A (en) |
CA (1) | CA2267954C (en) |
CZ (1) | CZ289022B6 (en) |
DE (1) | DE69624848T2 (en) |
WO (1) | WO1998016906A1 (en) |
Families Citing this family (43)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7630895B2 (en) * | 2000-01-21 | 2009-12-08 | At&T Intellectual Property I, L.P. | Speaker verification method |
US6847717B1 (en) | 1997-05-27 | 2005-01-25 | Jbc Knowledge Ventures, L.P. | Method of accessing a dial-up service |
US9978373B2 (en) | 1997-05-27 | 2018-05-22 | Nuance Communications, Inc. | Method of accessing a dial-up service |
IE980941A1 (en) * | 1998-11-16 | 2000-05-17 | Buy Tel Innovations Ltd | A transaction processings system |
EP1022888B1 (en) * | 1999-01-20 | 2005-05-25 | Swisscom Fixnet AG | Method or system to secure the access to telecommunication services in a network |
EP1039769B1 (en) * | 1999-03-24 | 2012-05-09 | Hewlett-Packard Development Company, L.P. | A method and apparatus for getting acccess to a digital mobile phone |
EP1039770B1 (en) * | 1999-03-24 | 2011-05-11 | Hewlett-Packard Development Company, L.P. | A method and apparatus for getting access to a mobile communication apparatus, particularly digital mobile phone |
US6691089B1 (en) * | 1999-09-30 | 2004-02-10 | Mindspeed Technologies Inc. | User configurable levels of security for a speaker verification system |
DE19952049A1 (en) * | 1999-10-28 | 2001-05-10 | Siemens Ag | Method and arrangement for verifying a speaker using a computer |
DE10000973A1 (en) | 2000-01-06 | 2001-07-12 | Deutsche Telekom Ag | Voice-controlled identification involves permitting clear identification once user has been recognized, thereby enabling user to be addressed personally |
DE10012925A1 (en) * | 2000-03-16 | 2001-10-18 | Deutsche Telekom Ag | Procedure for verifying a caller |
DE10039668A1 (en) * | 2000-08-14 | 2002-02-28 | Mettler Toledo Gmbh | Measuring equipment with user profile |
DE10111393A1 (en) * | 2001-03-09 | 2002-09-26 | Rexroth Indramat Gmbh | Remote control programming procedures |
JP3982195B2 (en) * | 2001-03-28 | 2007-09-26 | オムロン株式会社 | Target person verification device and target person verification system |
US20020174347A1 (en) | 2001-05-18 | 2002-11-21 | Imprivata, Inc. | Authentication with variable biometric templates |
US20020194003A1 (en) * | 2001-06-05 | 2002-12-19 | Mozer Todd F. | Client-server security system and method |
US7245716B2 (en) | 2001-12-12 | 2007-07-17 | International Business Machines Corporation | Controlling hold queue position adjustment |
US9088645B2 (en) * | 2001-12-12 | 2015-07-21 | International Business Machines Corporation | Intermediary device initiated caller identification |
US7167551B2 (en) * | 2001-12-12 | 2007-01-23 | International Business Machines Corporation | Intermediary device based callee identification |
US20030108159A1 (en) * | 2001-12-12 | 2003-06-12 | International Business Machines Corporation | Destination device based callee identification |
US20030108163A1 (en) * | 2001-12-12 | 2003-06-12 | International Business Machines Corporation | Origin device based caller identification |
US7443970B2 (en) * | 2001-12-17 | 2008-10-28 | International Business Machines Corporation | Logging calls according to call context |
US20030163739A1 (en) * | 2002-02-28 | 2003-08-28 | Armington John Phillip | Robust multi-factor authentication for secure application environments |
US20040010408A1 (en) * | 2002-07-10 | 2004-01-15 | Mani Babu V. | Method and system for identifying a caller based on voice recognition |
US7660880B2 (en) | 2003-03-21 | 2010-02-09 | Imprivata, Inc. | System and method for automated login |
US7941849B2 (en) | 2003-03-21 | 2011-05-10 | Imprivata, Inc. | System and method for audit tracking |
US20040187029A1 (en) | 2003-03-21 | 2004-09-23 | Ting David M. T. | System and method for data and request filtering |
PL2337016T3 (en) * | 2004-04-30 | 2018-07-31 | Idhl Holdings Inc | Free space pointing devices with tilt compensation and improved usability |
KR100985364B1 (en) | 2004-04-30 | 2010-10-04 | 힐크레스트 래보래토리스, 인크. | Free space pointing device and method |
US8629836B2 (en) | 2004-04-30 | 2014-01-14 | Hillcrest Laboratories, Inc. | 3D pointing devices with orientation compensation and improved usability |
WO2006058129A2 (en) | 2004-11-23 | 2006-06-01 | Hillcrest Laboratories, Inc. | Semantic gaming and application transformation |
JP4825431B2 (en) * | 2005-03-08 | 2011-11-30 | パナソニック株式会社 | Access control device |
US8374324B2 (en) | 2005-06-02 | 2013-02-12 | At&T Intellectual Property I, L.P. | Methods of using biometric data in a phone system and apparatuses to perform the methods |
CN1905445B (en) | 2005-07-27 | 2012-02-15 | 国际商业机器公司 | System and method of speech identification using mobile speech identification card |
WO2008018136A1 (en) * | 2006-08-10 | 2008-02-14 | Pioneer Corporation | Speaker recognizing device, speaker recognizing method, etc. |
EP2034432A1 (en) * | 2007-09-05 | 2009-03-11 | Deutsche Post AG | Method and Device for performing Transactions |
US8260350B2 (en) * | 2008-10-24 | 2012-09-04 | International Business Machines Corporation | Embedded biometrics in telecommunication devices for feature extraction and context identification |
US8645136B2 (en) * | 2010-07-20 | 2014-02-04 | Intellisist, Inc. | System and method for efficiently reducing transcription error using hybrid voice transcription |
US9318114B2 (en) * | 2010-11-24 | 2016-04-19 | At&T Intellectual Property I, L.P. | System and method for generating challenge utterances for speaker verification |
US8818810B2 (en) | 2011-12-29 | 2014-08-26 | Robert Bosch Gmbh | Speaker verification in a health monitoring system |
SG11201504186UA (en) | 2012-12-19 | 2015-07-30 | Visa Int Service Ass | System and method for voice authentication |
US10044710B2 (en) * | 2016-02-22 | 2018-08-07 | Bpip Limited Liability Company | Device and method for validating a user using an intelligent voice print |
CN114220209B (en) * | 2022-02-22 | 2022-05-13 | 深圳灏鹏科技有限公司 | Hotel access control method, hotel access control device, hotel access control equipment and storage medium |
Family Cites Families (15)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5202929A (en) * | 1979-09-24 | 1993-04-13 | Lemelson Jerome H | Data system and method |
US5375161A (en) * | 1984-09-14 | 1994-12-20 | Accessline Technologies, Inc. | Telephone control system with branch routing |
US5548647A (en) * | 1987-04-03 | 1996-08-20 | Texas Instruments Incorporated | Fixed text speaker verification method and apparatus |
US4827518A (en) * | 1987-08-06 | 1989-05-02 | Bell Communications Research, Inc. | Speaker verification system using integrated circuit cards |
CA2040724A1 (en) * | 1990-04-25 | 1991-10-26 | James L. Flanagan | Voice password controlled computer security system |
US5517558A (en) | 1990-05-15 | 1996-05-14 | Voice Control Systems, Inc. | Voice-controlled account access over a telephone network |
US5365574A (en) * | 1990-05-15 | 1994-11-15 | Vcs Industries, Inc. | Telephone network voice recognition and verification using selectively-adjustable signal thresholds |
US5127043A (en) * | 1990-05-15 | 1992-06-30 | Vcs Industries, Inc. | Simultaneous speaker-independent voice recognition and verification over a telephone network |
US5274695A (en) * | 1991-01-11 | 1993-12-28 | U.S. Sprint Communications Company Limited Partnership | System for verifying the identity of a caller in a telecommunications network |
US5430827A (en) * | 1993-04-23 | 1995-07-04 | At&T Corp. | Password verification system |
US5806040A (en) * | 1994-01-04 | 1998-09-08 | Itt Corporation | Speed controlled telephone credit card verification system |
US5907597A (en) * | 1994-08-05 | 1999-05-25 | Smart Tone Authentication, Inc. | Method and system for the secure communication of data |
US5592531A (en) * | 1995-04-03 | 1997-01-07 | Motorola, Inc. | Selective call communication system having adaptive voice data entry |
US5726978A (en) * | 1995-06-22 | 1998-03-10 | Telefonaktiebolaget L M Ericsson Publ. | Adaptive channel allocation in a frequency division multiplexed system |
US5913196A (en) * | 1997-11-17 | 1999-06-15 | Talmor; Rita | System and method for establishing identity of a speaker |
-
1996
- 1996-10-15 JP JP51793498A patent/JP3904608B2/en not_active Expired - Fee Related
- 1996-10-15 DE DE69624848T patent/DE69624848T2/en not_active Expired - Lifetime
- 1996-10-15 AT AT96934692T patent/ATE227868T1/en not_active IP Right Cessation
- 1996-10-15 CA CA002267954A patent/CA2267954C/en not_active Expired - Fee Related
- 1996-10-15 EP EP96934692A patent/EP0932885B1/en not_active Expired - Lifetime
- 1996-10-15 AU AU72930/96A patent/AU7293096A/en not_active Abandoned
- 1996-10-15 WO PCT/EP1996/004480 patent/WO1998016906A1/en active IP Right Grant
- 1996-10-15 US US09/284,140 patent/US6556127B1/en not_active Expired - Lifetime
- 1996-10-15 CZ CZ19991257A patent/CZ289022B6/en not_active IP Right Cessation
Also Published As
Publication number | Publication date |
---|---|
CZ289022B6 (en) | 2001-10-17 |
EP0932885A1 (en) | 1999-08-04 |
WO1998016906A1 (en) | 1998-04-23 |
JP3904608B2 (en) | 2007-04-11 |
CZ125799A3 (en) | 1999-08-11 |
AU7293096A (en) | 1998-05-11 |
JP2001503156A (en) | 2001-03-06 |
US6556127B1 (en) | 2003-04-29 |
EP0932885B1 (en) | 2002-11-13 |
ATE227868T1 (en) | 2002-11-15 |
DE69624848T2 (en) | 2003-10-09 |
DE69624848D1 (en) | 2002-12-19 |
CA2267954A1 (en) | 1998-04-23 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CA2267954C (en) | Speaker verification method | |
US10122712B2 (en) | Voice over IP based biometric authentication | |
US5717743A (en) | Transparent telephone access system using voice authorization | |
US5274695A (en) | System for verifying the identity of a caller in a telecommunications network | |
US7360239B2 (en) | Biometric multimodal centralized authentication service | |
US8626513B2 (en) | Centralized biometric authentication | |
US5517558A (en) | Voice-controlled account access over a telephone network | |
US6119084A (en) | Adaptive speaker verification apparatus and method including alternative access control | |
US6973426B1 (en) | Method and apparatus for performing speaker verification based on speaker independent recognition of commands | |
US5315649A (en) | Toll call telephone service center | |
EP0451695A2 (en) | Method and apparatus for securing access to telecommunications based assets and services | |
US8406383B2 (en) | Voice authentication for call control | |
EP0451693A2 (en) | Method and apparatus for providing integrated access to telecommunication based services | |
EP0453831A2 (en) | User interfaces for telecommunications based services | |
EP1164576B1 (en) | Speaker authentication method and system from speech models | |
US20230120373A1 (en) | Strong authentication of a user of a communication terminal | |
JP2004514308A (en) | How to identify a caller under a telephone service operator |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
EEER | Examination request | ||
MKLA | Lapsed |