Conversational Biometrics Group - HOPE Demo
HTML Online Parsing Entity

Conversational Biometrics Travel System

The HOPE travel system provides a telephony speech interface to web content, including travel information (* and other types of web content such as IBM blue pages. The hope system invokes a conversational biometrics session in order to verify the caller. Subsequently, upon verification of the user it accesses travel web content using the HOPE server, allowing the user to navigate through travel reservation data. The information includes flights, hotel, rental car information, and a live weather forecast for the trip destinations. The HOPE server is also used for personalization features: the user's data is retrieved from IBM Blue Pages Intranet, and used to generate a personalized greeting including the user's name and interests.

Conversational speech biometrics combines two sources for authentication: 1) physical speech biometrics (the voice-print) and: 2) user knowledge (e.g.,passwords and personal information). The combination of the two information sources increases the security and reliability and provides a flexible framework for various authentication scenarios so as to maximize user convenience. Technologies utilized to enable conversational speech biometrics include acoustic speaker recognition, speech recognition, natural language understanding, and dialog management. In the HOPE application, the verification consists of one or several short interviews involving randomly asked authentication questions and an acoustic voice-print check. The length of a session depends on the correctness of the answers and the estimated voice-print confidence. Users can enroll into the system via an HTML-form and a telephone server. Measured on real data, casual impostors are falsely accepted by the system in less than 0.00001 % of the cases, balanced by a 3% false rejection of the clients.

*) Sabre and Virtually There are trademarks of an affiliate of Sabre Inc.

Scenario 1: An authentic user accesses his account (user="ran"). The user calls and gains access to the travel reservation data after being asked two questions. Both the answers to the questions and the voiceprint match the user's enrollment data. This example also includes navigation through the travel data once access is granted. | SCENARIO 1 WAV
True User
Scenario2: An imposter is trying to intrude another user's account. Besides the voice mismatch, the imposter has no knowledge available to answer the biometrics questions correctly. | SCENARIO 2 WAV
Uninformed Impostor
Scenario 3: A (well informed) imposter is trying to break into another user's account. In this scenario, the imposter is in possession of the complete list with correct answers to the biometrics questions of the true user. In praxis, capturing the complete question pool -- for example by eaves-dropping -- is made difficult by randomizing the interviews and asking different questions in every consecutive session. | SCENARIO 3 WAV
Informed Impostor