Vocal identification of devices in a home environment

• US 6,052,666 A
• Filed: 10/09/1996
• Issued: 04/18/2000
• Est. Priority Date: 11/06/1995
• Status: Expired due to Term

• Current Assignee: Thomson Multimedia Incorporated (Vantiva SA)
• Original Assignee: Thomson Multimedia Incorporated (Vantiva SA)
• Inventors: Diehl, Eric, Corda, Gerard
• Primary Examiner(s): Dorvil, Richemond
• Application Number:
  US08/728,488
• Time in Patent Office: 1,287 Days
• Field of Search: 704/270, 704/275, 704/272, 704/258, 704/274
• US Class Current: 704/275
• CPC Class Codes: B29L 2031/286 Cutlery
• CPC Class Codes: G10L 13/00 Speech synthesis; Text to speech systems

Abstract

A speech based man-machine communication system is given, comprising more than one controllable device provided with speech synthesis function. Each of the devices in question is provided with its own unique voice pattern. The devices are connected via a bus, so that a central authority handles all the requests from the user. Because the user uses his natural language, commands can be ambiguous. Therefore an algorithm for handling ambiguous situations is provided.

Claims

1. A method for controlling a plurality of consumer electronic target devices interconnected via a digital bus to a central authority device, said central authority performing the steps of:

receiving a speech command from a user;

evaluating said received speech command to determine whether one of said target devices is identified;

issuing, in response to said identified target device, a corresponding command from said central authority to said target device;

evaluating said received speech command to determine whether said received speech command is unique to a specific target device;

issuing, in response to said unique received speech command, a corresponding command from said central device to said specific target device; and

thereafter

generating, in response to said speech command not identifying said target device and not being unique to one of said target devices, a list of target devices capable of being controlled by said received speech command.

2. The method according to claim 1 wherein each of said target devices identified on said list performs the step of:

generating uniquely identifiable speech information in response to said received speech command.

3. The method of claim 2 wherein said uniquely identifiable speech information comprises a feminine accent.

4. The method of claim 2 wherein said uniquely identifiable speech information comprises a masculine accent.

5. The method according to claim 1, wherein the step of generating said list of target devices is generated according to a probability approach.

6. A method for controlling a plurality of consumer electronic target devices interconnected via a digital bus to a central authority device, said central authority performing the steps of:

evaluating a speech command received from a user to determine whether said speech command is unique to a specific target device;

issuing, in response to said unique speech command, a corresponding command from said central device to said specific target device; and

thereafter

indicating, in response to said speech command not being unique to one of said target devices, which of said target devices is capable of being controlled by said speech command.