Data processing: speech signal processing – linguistics – language – Speech signal processing – Application
Reexamination Certificate
1996-09-09
2001-09-18
Dorvil, Richemond (Department: 2641)
Data processing: speech signal processing, linguistics, language
Speech signal processing
Application
C704S275000
Reexamination Certificate
active
06292782
ABSTRACT:
TECHNICAL FIELD
The present invention relates to secure data transmission, and more particularly to a method and apparatus authorizing data transmission over a computer network environment utilizing a client/server architecture.
BACKGROUND OF THE INVENTION
The current expansion of computer network systems and linked architectures, such as the Internet, have greatly increased the opportunity for carrying out transactions through networked computers. Unfortunately, the performance of transactions via networked computers requires the transmission of sensitive data (credit card numbers, PIN numbers, passwords, etc.) over the communications network interconnecting the computers. A communications network is for the most part a nonsecure transmission environment that is subject to access by unauthorized third parties. An unauthorized party is able to get a hold of sensitive information by unlawfully monitoring computer communications over the communications network. This of course can be greatly damaging to an individual or business. Also, transfer of data over a communications network does not provide an easy way to ensure a party is authorized to transmit or receive sensitive data.
One current method for providing secure transmission of sensitive data in a computer network environment relies on encryption of the data prior to transmission. Unfortunately, newly devised decryption methods and faster computer systems continually make it easier for unauthorized third parties to crack any encryption code, thus rendering sensitive data vulnerable to unauthorized attack. Once the sensitive data has been decrypted by an unauthorized individual, this party may now use this sensitive data without authorization to make purchases or carry out any number of unauthorized transactions. Since many current encryption methods have no mechanism for verifying the identity of the person submitting the sensitive information, the unauthorized individual may continue their unlawful activities for a substantial period of time.
Other current systems providing authentication require that additional hardware be purchased by the user and an authentication card. The user must insert the authentication card into an associated card reader to access sensitive data. If the card is illegally obtained by an unauthorized individual, this person may still access the sensitive data. Thus, a system enabling secure transmission of sensitive data and verification of a sender's identity would greatly benefit the expanded use of transactions occurring over networked computer systems.
BRIEF SUMMARY OF THE INVENTION
The present invention overcomes the foregoing and other problems with a system and method using speech recognition and verification engines to enable secure data transmissions between networked computer systems. The preferred embodiment of this invention utilizes a client/server architecture wherein a request for a particular transaction by a user at the client unit causes the server unit to prompt the user for a spoken identifier, such as a credit card number, PIN number, password, etc. The user speaks an identifier into a microphone connected to the client unit and the spoken data comprising the identifier is converted into speech feature data. This conversion is carried out by a locally resident software application or may be done by an application or plug-in applet that has been downloaded from the server unit in response to the requested transaction.
The speech feature data is transmitted over a computer communications link from the client unit to the server unit for further processing. Optionally, the speech feature data may be secured by additional state of the art processes, such as encryption, before transmission of the speech feature data to the server unit. A speech recognition engine located at the server unit uses the speech feature data to identify and confirm the spoken identifier entered by the user. The speech feature data is then further processed by a speech verification engine to confirm that the user who entered the spoken identifier is in fact the user associated with the spoken identifier and is authorized to perform a requested transaction.
Additionally, the speech feature data for the identifier may be compared to previously transmitted versions of the speech feature data for the identifier to determine if it matches any of the previously transmitted versions. When an exact match exists, the transmitted speech feature data is marked as suspect so that further approval steps may be taken. If the spoken identifier is recognized and verified as being associated with the user entering the identifier and no questions arise due to an exact match with previously transmitted data, the transaction request is completed.
The foregoing has outlined some of the more pertinent aspects of the present invention. These aspects should be construed to be merely illustrative of some of the more prominent features and applications of the invention. Many other beneficial results can be attained by applying the disclosed invention in a modified manner as will be described. Accordingly, other aspects and a fuller understanding of the invention may be had by referring to the following Detailed Description of the preferred embodiment.
REFERENCES:
patent: 5127043 (1992-06-01), Hunt et al.
patent: 5297194 (1994-03-01), Hunt et al.
patent: 5343529 (1994-08-01), Goldfine et al.
patent: 5465387 (1995-11-01), Mukherjee
patent: 5517558 (1996-05-01), Schalk
patent: 5610981 (1997-03-01), Mooney et al.
patent: 5761329 (1998-06-01), Chen et al.
patent: 5805674 (1998-09-01), Anderson, Jr.
patent: 5806040 (1998-09-01), Vensko
patent: 5838682 (1998-11-01), Dekelbaum et al.
patent: 5901203 (1999-05-01), Morganstein et al.
patent: 5907597 (1999-05-01), Mark
patent: 5915001 (1999-06-01), Uppaluru
Dorvil Richemond
Philips Electronics North America Corp.
Piotrowski Daniel J.
LandOfFree
Speech recognition and verification system enabling... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Speech recognition and verification system enabling..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Speech recognition and verification system enabling... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2446917