Summary generating device, summary generating method, and computer program product

• US 10,540,987 B2
• Filed: 01/26/2017
• Issued: 01/21/2020
• Est. Priority Date: 03/17/2016
• Status: Active Grant

• Current Assignee: Kabushiki Kaisha Toshiba (Toshiba Corporation)
• Original Assignee: Kabushiki Kaisha Toshiba (Toshiba Corporation)
• Inventors: Fume, Kosei, Ashikawa, Taira, Ashikawa, Masayuki, Masuko, Takashi
• Primary Examiner(s): Washburn, Daniel C
• Assistant Examiner(s): Ogunbiyi, Oluwadamilola M
• Application Number:
  US15/416,973
• Publication Number:
  US 20170270949A1
• Time in Patent Office: 1,090 Days
• CPC Class Codes: G06F 16/345 Summarisation for human users
• CPC Class Codes: G06F 40/103 Formatting, i.e. changing of presentation of documents automatic justification G06F40/189; automatic line break hyphenation G06F40/191
• CPC Class Codes: G06F 40/268 Morphological analysis
• CPC Class Codes: G10L 15/02 Feature extraction for speech recognition; Selection of recognition unit
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence
• CPC Class Codes: G10L 2015/221 Announcement of recognition results
• CPC Class Codes: G10L 21/10 Transforming into visible information
• CPC Class Codes: G10L 25/78 Detection of presence or absence of voice signals switching of direction of transmission by voice frequency in two-way loud-speaking telephone systems H04M9/10

Abstract

A summary generating device includes a featural script extracting unit, a segment candidate generating unit, and a structuring estimating unit. The featural script extracting unit extracts featural script information of the words included in text information. Based on the extracted feature script information, the segment candidate generating unit generates candidates of segments that represent the constitutional units for the display purpose. Based on the generated candidates of segments and based on an estimation model for structuring, the structuring estimating unit estimates structure information containing information ranging from information of a comprehensive structure level to information of a local structure level.

Claims

1. A summary generating device comprising:

a memory; and

one or more processors configured to function as a featural script extracting unit, a segment candidate generating unit and a structuring estimating unit, wherein

the featural script extracting unit extracts featural script information from text information, the featural script information including at least one of a part-of-speech and metadata of a logical element included in the text information, property information of the part-of-speech, and system information related to the summary generating system;

the segment candidate generating unit generates, based at least in part on the extracted featural script information, candidates of segments representing constitutional units for display, and performs ordering of sets of the featural script information; and

a structuring estimating unit estimates structure information comprising structural elements based at least in part on the generated candidates of segments and a structure estimation model, the structural elements comprising a comprehensive structural element that has a comprehensive structure level and a local structural element that has a local structure level, the comprehensive structure level representing a broadest structural element within a hierarchy of the structural elements and the local level representing a narrowest structural element within the hierarchy of the structural elements, the structure estimation model including learning data obtained by learning display formats suitable for displaying and summary results edited or decided prior to the estimation,

wherein, the structuring estimating unit presents a structural element of the structural elements having a likelihood that is higher than another structure element of the structural elements, the likelihood being the likelihood with respect to the estimation model.

2. The summary generating device according to claim 1, wherein

the one or more processors is configured to further function as a voice recognizing unit,

the voice recognizing unit performs voice recognition with respect to voice data and generates the text information based on the voice recognition of the voice data, and

the featural script extracting unit extracts featural script of words included in the generated text information.

3. The summary generating device according to claim 2, wherein

the voice recognizing unit generates the text information to include character information of utterances and timings of the utterances, and extracts audio features of the voice data, and

the featural script extracting unit extracts timings corresponding to the character information and extracts the audio features as surrounding information with respect to the words.

4. The summary generating device according to claim 1, wherein

the one or more processors is configured to further function as a display format converting unit, and

the display format converting unit converts the estimated structure information into a display format for viewing.

5. The summary generating device according to claim 4, wherein the display format converting unit changes the display format according to a user instruction.

6. The summary generating device according to claim 1, wherein the segment candidate generating unit performs the ordering of the sets of the featural script information based at least in part on at least one of a character length of each featural script or an inclusion number of specific elements.

7. A summary generating method implemented by a summary generating device including a memory and one or more processors configured to function as a featural script extracting unit, a segment candidate generating unit and a structuring estimating unit, the summary generating method comprising:

extracting, by the featural script extracting unit, featural script information of words from text information, the featural script information including at least one of a part-of-speech and metadata of a logical element included in the text information, property information of the part-of-speech, and system information related to the summary generating system;

generating, by the segment candidate generating unit, based at least in part on the extracted featural script information;

estimating, by the structuring estimating unit, based at least in part on the generated candidates of segments and a structure estimation model, structure information comprising structural elements comprising a comprehensive structural element that has a comprehensive structure level and a local structural element that has a local structure level, the comprehensive structure level representing a broadest structural element within a hierarchy of the structural elements the local level representing a narrowest structural element within the hierarchy of the structural elements, the structure estimation model including learning data obtained by learning display formats suitable for displaying and summary results edited or decided prior to the estimation; and

performing ordering of sets of the featural script information,

wherein, the structuring estimating unit presents a structural element of the structural elements having a likelihood that is higher than another structure element of the structural elements, the likelihood being the likelihood with respect to the estimation model.

8. The summary generating method according to claim 7, wherein said performing ordering of the sets of the featural script information comprises performing ordering of the sets of the featural script information based at least in part on at least one of a character length of each featural script or an inclusion number of specific elements.

9. A computer program product comprising a non-transitory computer readable medium that comprises a summary generating program, wherein the summary generating program, when executed by a computer, causes the computer to perform:

extracting featural script information of words from text information, the featural script information including at least one of a part-of-speech and metadata of a logical element included in the text information, property information of the part-of-speech, and system information related to the summary generating system;

generating, based at least in part on the extracted featural script information, candidates of segments representing constitutional units for display;

estimating, based at least in part on the generated candidates of segments and a structure estimation model, structure information comprising structure elements comprising a comprehensive structural element that has a comprehensive structure level and a local structural element that has a local structure level, the comprehensive structure level representing a broadest structural element within a hierarchy of the structural elements and the local level representing a narrowest structural element within the hierarchy of the structural elements, the structure estimation model including learning data obtained by learning display formats suitable for displaying and summary results edited or decided prior to the estimation; and

performing ordering of sets of the featural script information,

wherein, the structuring estimating unit presents a structural element of the structural elements having a likelihood that is higher than another structure element of the structural elements, the likelihood being the likelihood with respect to the estimation model.

10. The summary generating program according to claim 9, wherein said performing ordering of the sets of the featural script information comprises performing ordering of the sets of the featural script information based at least in part on at least one of a character length of each featural script or an inclusion number of specific elements.