IP Library › Granted Patent US 7,801,508
Granted Patent B2
US 7,801,508 · App. 10/987,963 · Granted Sep 21, 2010

Method for authentication of a user on the basis of his/her voice profile

View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 7,801,508
App. No.
10/987,963
Filed
Nov 12, 2004
Granted
Sep 21, 2010
Kind
B2
Art Unit
2617
USPC
455/411
Abstract

Voice recording used for authentication is transmitted to the voice portal of the service provider not by way of the voice channel, but rather by way of a data channel. In this connection, the voice recording is sent not synchronous to speech and subject to loss, but rather asynchronously and loss-free, in an extra data package (for example advantageously as a Multimedia Messaging Service (MMS) data package). For this purpose, the resources that are available as standard items in most mobile phone terminals, such as digital voice recording and MMS transmission, can be utilized. Preferably, the subscriber's identification module or SIM card in the mobile phone terminal can store and implement the corresponding control program.

Claims (12)

1. A method for authentication of a user by a voice profile of the user in the course of a speech dialog, which comprises the following steps:

(a) establishing a voice connection via a voice channel of a mobile network;

(b) recording a voice sample of the user by means of a microphone of a mobile terminal;

(c) transmitting to a server, asynchronously to speech and parallel in time with the course of the speech dialog, the voice sample as at least one separate data package via a data channel of the mobile network, wherein reduced bandwidths of the data channel are compensated by extending transmission time of the voice sample;

(d) checking the voice sample by means of the server, wherein the voice sample is compared with a recorded voice profile to thereby authenticate the user.

2. The method as recited in claim 1 , wherein the at least one separate data package is generated using a speech recognition function that is implemented at the mobile terminal.

3. The method as recited in claim 1 , wherein the voice sample is sent as a Multimedia Messaging Service (MMS) message.

4. The method as recited in claim 1 , wherein recording and transmission of the voice sample is controlled by means of a subscriber's identification module (SIM) card of the terminal.

5. The method as recited in claim 4 , wherein recording and transmission of the voice sample begins after a service has authenticated itself to the SIM card of the terminal, and has instructed that the voice sample be recorded and transmitted.

6. The method as recited in claim 1 , wherein the at least one data package is digitally signed before transmission.

7. The method as recited in claim 1 , wherein sending of the voice sample as a data package takes place parallel in time with a continued dialog between the server and the user.

8. The method as recited in claim 1 , wherein the user is requested by the server, for the purpose of recording the voice sample, to make a statement that is exclusively selected from a number of possible statements.