Data processing: speech signal processing – linguistics – language – Speech signal processing – Recognition
Reexamination Certificate
1999-09-30
2004-02-10
Dorvil, Richemond (Department: 2654)
Data processing: speech signal processing, linguistics, language
Speech signal processing
Recognition
C704S250000, C704S273000
Reexamination Certificate
active
06691089
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates generally to a system and method for providing a speaker verification system with user selectable levels of security.
2. Related Art
The increased use of consumer electronic devices and computer controlled remote services has heightened concerns over security issues. One of the primary security concern is the risk of access by unauthorized users. In order to safeguard against unauthorized use, passwords and/or user identification codes are generally provided.
Therefore, users of these devices and services are required to memorize and maintain a variety of passwords and/or user identification codes (user IDs) to maintain security. For example, user IDs and/or passwords (hereinafter separately and/or collectively referred to as passwords) are generally required when using ATM cards, credit cards, telephone calling cards, bank accounts, residential security systems, personal computer systems, remote computer services, voice mail systems, pagers, cellular telephones and personal digital assistants (PDAs).
It has become apparent that users are finding it difficult and inconvenient to memorize and maintain these passwords. This is especially true for users of multiple devices and/or services. The consequence of this inconvenience almost always results in some sort of breach of security. For example, rather than commit multiple passwords to memory, many users will write them down and thereby increase the risk of misappropriation. In another example, this inconvenience causes users to avoid setting up optional passwords altogether. In yet another example, users tend to use trivial passwords, such as their birth dates that are easily compromised. In addition, many users tend to use the same passwords across multiple systems. Thus, if one password is compromised, all of them are.
Therefore, to alleviate this increasing and prevailing problem, what is needed is a system and method for maintaining a high level of security that avoids the inconveniences found in current password authorization systems.
SUMMARY OF THE INVENTION
Accordingly, the present invention is directed toward a system and method for providing a security system that avoids inconveniences and problems found in conventional systems. The system and method of the present invention uses a text-prompted speaker verification system to accept randomly generated verbal passwords from users. The system and method of the present invention can be used with any type of electronic device and/or any type of computer controlled local or remote automated service. In fact, the present invention can be used in any system where passwords may be utilized.
The present invention prompts the user (either verbally or textually) for a multiple-digit (or multiple-word) password used for gaining access to the system. The number of digits or words used for each password is defined by the system in accordance with a preferred level of user security. In addition, the level of training required by the system is user configurable and based on the desired level of security.
Further, the ease of system access is user configurable and based on a desired level of security. Specifically, in accordance with a preferred embodiment of the present invention, the types of errors that may be generated by the system are user configurable. This is accomplished by allowing the user to adjust the acceptable frequency of errors between the two types of possible errors, false accepts and false rejects.
Once a particular level of security is defined by a user, the user gains access to the system by uttering a randomly generated password as prompted by the system. The number of words or digits used for the password, the amount of user training required, and the acceptable error frequency and type, are all configured by the user based on a desired level of security as defined by the user for each particular application.
REFERENCES:
patent: 5724481 (1998-03-01), Garberg et al.
patent: 5774858 (1998-06-01), Taubkin et al.
patent: 5805674 (1998-09-01), Anderson, Jr.
patent: 5913196 (1999-06-01), Talmor et al.
patent: 5940799 (1999-08-01), Bruckert et al.
patent: 5960397 (1999-09-01), Rahim
patent: 5963908 (1999-10-01), Chadha
patent: 6088669 (2000-07-01), Maes
patent: 6107935 (2000-08-01), Comerford et al.
patent: 6119084 (2000-09-01), Roberts et al.
patent: 6182037 (2001-01-01), Maes
patent: 6253181 (2001-06-01), Junqua
patent: 6272463 (2001-08-01), Lapere
patent: 6389393 (2002-05-01), Gong
patent: 6556127 (2003-04-01), Moser et al.
patent: 0 892 387 (1999-01-01), None
patent: WO 99/42992 (1999-08-01), None
Vloeberghs et al.; “La Reconnaissance Du Locuteur Sur Des Liasons Par Vocodeurs”; Revue HF, Soc. Belge Des Ing. Des Telecomm. & D'Electronique; Ophain, BE; vol. 17, No. Jan. 2, 2003; 1993; pp. 119-127.
Assaleh Khaled
Su Huan-yu
LandOfFree
User configurable levels of security for a speaker... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with User configurable levels of security for a speaker..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and User configurable levels of security for a speaker... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3328021