Information providing system

• US 9,978,368 B2
• Filed: 09/16/2014
• Issued: 05/22/2018
• Est. Priority Date: 09/16/2014
• Status: Active Grant

• Current Assignee: Mitsubishi Electric Corporation
• Original Assignee: Mitsubishi Electric Corporation
• Inventors: Osawa, Masanobu, Takei, Takumi, Furumoto, Yuki, Narita, Tomohiro, Saito, Tatsuhiko
• Primary Examiner(s): ALBERTALLI, BRIAN LOUIS
• Application Number:
  US15/315,506
• Publication Number:
  US 20170200448A1
• Time in Patent Office: 1,344 Days
• CPC Class Codes: G10L 15/10 using distance or distortion measures between unknown speech and reference templates
• CPC Class Codes: G10L 15/1815 Semantic context, e.g. disambiguation of the recognition hypotheses based on word meaning
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence
• CPC Class Codes: G10L 2015/088 Word spotting
• CPC Class Codes: G10L 2015/223 Execution procedure of a spoken command

Abstract

Because an information providing system according to the present invention generates recognition object words for selecting one headline from a plurality of headlines in such a way that the recognition object words do not overlap one another, the information providing system can select a headline uniquely through the user'"'"'s utterance, and present information (content) corresponding to the headline selected thereby, and therefore the convenience to the user is improved.

Claims

1. An information providing system that presents a plurality of headlines, selects one of said plurality of headlines, and presents a piece of content corresponding to the headline selected thereby, said information providing system comprising:

an extractor to acquire said plurality of headlines, and to extract recognition object word candidates from said headlines acquired thereby, respectively;

an object word determiner to acquire the recognition object word candidates of said headlines from said extractor, and to generate recognition object words of said headlines in accordance with said acquired recognition object word candidates of said headlines; and

a controller to, when a voice recognition result acquired from a voice matches a recognition object word generated by said object word determiner, issue a command to present a piece of content corresponding to a headline from which the recognition object word matching said voice recognition result is determined,

wherein when recognition object word candidates overlapping each other exist in said acquired recognition object word candidates of said headlines, said object word determiner dynamically generates the recognition object words of said headlines in such a way that the recognition object words of said headlines differ from one another.

2. The information providing system according to claim 1, wherein when recognition object word candidates overlapping each other exist in said acquired recognition object word candidates of said headlines, said object word determiner processes at least one of said recognition object word candidates overlapping each other, to dynamically generate the recognition object words of said headlines in such a way that the recognition object words differ from one another.

3. The information providing system according to claim 2, wherein said object word determiner identifies either a presentation position or a position in presentation order of a headline including one of said recognition object word candidates overlapping each other, and processes said one of the recognition object word candidates in accordance with either said presentation position or said position in the presentation order, which is identified thereby.

4. The information providing system according to claim 2, wherein said object word determiner processes one of said recognition object word candidates overlapping each other by adding either another word or a word string to said one of the recognition object word candidates.

5. The information providing system according to claim 1, wherein when recognition object word candidates overlapping each other exist in said acquired recognition object word candidates of the headlines, said object word determiner replaces at least one of said recognition object word candidates overlapping each other with another recognition object word candidate.

6. The information providing system according to claim 5, wherein said object word determiner analyzes a meaning of a headline including the one of said recognition object word candidates overlapping each other, determines a term corresponding to said meaning as the other recognition object word candidate, and replaces at least the one of said recognition object word candidates overlapping each other with the other recognition object word candidate.

7. The information providing system according to claim 5, wherein said object word determiner replaces the one of said recognition object word candidates overlapping each other with another recognition object word candidate included in a headline including the one of said recognition object word candidates.

8. The information providing system according to claim 1, wherein before dynamically generating the recognition object words of said headlines in such away that the recognition object words of said headlines differ from one another, said object word determiner determines said recognition object word candidates as the recognition object words of said headline, respectively, regardless of whether recognition object word candidates overlapping each other exist in said acquired recognition object word candidates of the headlines, and when the voice recognition result acquired from the voice matches a plurality of recognition object words determined by said object word determiner, said controller narrows down said headlines to two or more headlines from which said recognition object words identical to each other are determined, and commands said object word determiner to re-generate a recognition object word candidate different from said recognition object word candidates overlapping each other as a recognition object word for each of said headlines after narrowing-down.

9. The information providing system according to claim 1, wherein said object word determiner dynamically generates said recognition object words in such a way that each of said recognition object words differs from any operating command for equipment.