Apparatus and method of program classification using observed cues in the transcript information

• US 7,210,157 B2
• Filed: 12/18/2000
• Issued: 04/24/2007
• Est. Priority Date: 12/18/2000
• Status: Expired due to Fees

• Current Assignee: Koninklijke Philips Electronics N.V. (Koninklijke Philips N.V.)
• Original Assignee: Koninklijke Philips Electronics N.V. (Koninklijke Philips N.V.)
• Inventors: Devara, Kavitha
• Primary Examiner(s): BUI, KIEU OANH T
• Application Number:
  US09/739,476
• Publication Number:
  US 20020078452A1
• Time in Patent Office: 2,318 Days
• Field of Search: 725 18- 20, 725136-137
• US Class Current: 725/18
• CPC Class Codes: G06F 16/7844 using original textual content or text extracted from visual content or transcript of audio data
• CPC Class Codes: H04H 20/28 Arrangements for simultaneous broadcast of plural pieces of information
• CPC Class Codes: H04N 21/26603 for automatically generating descriptors from content, e.g. when it is not made available by its provider, using content analysis techniques
• CPC Class Codes: H04N 21/426 Internal components of the client ; Characteristics thereof H04N5/44 takes precedence
• CPC Class Codes: H04N 21/47 End-user applications

Abstract

A method for classification of a program includes receiving transcript information associated with the program, identifying at least one cue of a plurality of cues in the transcript information wherein each of the plurality of cues has associated therewith a type of program, correlating the at least one cue of the plurality of cues identified in the transcript information to the type of program, and classifying the program based on the correlation of the at least one cue of the plurality of cues identified in the transcript information.

Claims

1. A method for classification of a program comprising the steps of:

receiving an audio/video signal corresponding to the program;

determining transcript information associated with the program using the audio/video signal;

identifying at least one cue in the transcript information and an associated time of occurrence, each of the cues being associated with a type of program;

correlating the at least one cue identified in the transcript information to the type of program;

comparing the time of occurrence of two cues selected from the at least one identified cue and determining a proximity of occurrence of the two selected cues,

wherein if the proximity of occurrence is greater than a predetermined amount, the two selected cues are ignored in connection with determining the program classification; and

wherein if the proximity of occurrence is not greater than the predetermined amount, the two selected cues are utilized in connection with determining the program classification; and

classifying the program based on the proximity of occurrence of the two selected cues.

2. The method for classification of a program according to claim 1 further comprising the steps of:

receiving an audio/data/video signal which includes the transcript information.

3. The method for classification of a program according to claim 1, wherein the classification of the program is one of a news program, talk show, sports program, panel discussions, interviews, and situational comedy.

4. The method for classification of a program according to claim 1, wherein the transcript information comprises closed-captioned text.

5. Apparatus for classification of a program, comprising:

a receiver to receive an audio/data/video signal corresponding to the program;

a transcript information extractor for extracting transcript information associated with the program from the audio/data/video input signal;

a cue extractor for identifying at least one cue in the transcript information and an associated time of occurrence, each of the cues having associated therewith a type of program;

a knowledge database for correlating the at least one cue identified in the transcript information to the type of program;

a temporal database for comparing the time of occurrence of two selected cues, of the at least one cue to determine a proximity of occurrence of the two selected cues

wherein if the proximity of occurrence is greater than a predetermined amount, the two selected cues are ignored in connection with determining the program classification; and

wherein if the proximity of occurrence is not greater than the predetermined amount, the two selected cues are utilized in connection with determining the program classification; and

a classifier for classifying the program based on the proximity of occurrence.

6. The apparatus for classification of a program according to claim 5, wherein the classification of the program is one of a news program, talk show, sports program, panel discussions, interviews, and situational comedy.

7. The apparatus for classification of a program according to claim 5, wherein the transcript information comprises closed-captioned text.