Nippon Telegraph and Telephone Corporation

Setsuo Yamada

Takaaki Hasegawa

Yoshiaki Noda

Takashi Nakamura

Masayuki Sugizaki

Nobuaki Hiroshima

Kazuyuki Iso

A display data generation device includes an input unit that receives input of target data including a text sequence and annotation information corresponding to texts included in the text sequence, and a display preparation unit that determines, on the basis of the annotation information, annotation expression information indicating a background color of a display screen of a display device and a position and a range in which a corresponding background color is displayed for expressing correspondence relationship between the texts and the annotation information in a case where the display device displays the texts, and generates display data for causing the text sequence and the annotation information to be displayed according to a sequence in the text sequence, the display data being for causing the background color indicated by the annotation expression information to be displayed at the position and the range indicated by the annotation expression information.

Display data generation device, display data generation method, and display data generation program

Disclosed is a search result display device () comprising: a regard prediction unit () configured to predict, from a dialogue between a customer and a service person, a regard of the customer; a keyword extraction unit () configured to extract a keyword from the regard; and a display controller () configured to cause a display () to display the dialogue and a search result obtained from the database () with the keyword as a search query, wherein when a string has been designated by the service person, the display controller () causes the display () to display a search result obtained from the database () using a search query that incorporates the string, until a search result automatic update instruction is given by the service person.

Search result display device, search result display method, and non-transitory computer readable recording medium

The present invention allows appropriate acquisition of focus points in a dialogue. The focus point extraction devicecomprises: a dialogue scene predict unit () for predicting the dialogue scene of an utterance included in dialogue; an utterance type prediction sort unit () for predicting whether utterances are to be subjected to utterance type prediction; an utterance type predict unit () for predicting the utterance type of the utterance having been predicted as one to be subjected to the utterance type prediction; an utterance content extraction necessity predict unit () for predicting, in a case in which it has been predicted that the utterance belongs to any utterance type, whether a portion of the utterance is to be subjected, as utterance focus point information, to extraction or classification; and an utterance content extract unit () for: extracting or classifying the portion of the utterance from the utterance as the utterance focus point information, in a case in which it is predicted that a portion of the utterance is to be subjected to extraction or classification; and extracting or classifying the entirety of the utterance as the utterance focus point information, in a case in which it is predicted that a portion of the utterance is not to be subjected to extraction or classification.

Major point extraction device, major point extraction method, and non-transitory computer readable recording medium

The end-of-talk prediction device () of the present invention comprises: a divide unit () for dividing, using delimiter symbols indicating delimitations within segments, a string in which the utterance in the dialog has been text-converted by speech recognition, the delimiter symbols included in the result of the speech recognition; and an end-of-talk prediction unit () for predicting, using an end-of-talk prediction model (), whether the utterance corresponding to the divided string divided by the divide unit () is an end-of-talk utterance of the speaker.

End-of-talk prediction device, end-of-talk prediction method, and non-transitory computer readable recording medium

To improve prediction accuracy of utterance types in a dialog. A learning data generation device () according to the present invention comprises: a sort unit () configured to perform, based on information appended to utterances in a dialog amongst more than one speaker and that is indicative of a dialogue scene that is a scene in which the utterances in the dialog were made, sorting regarding whether the utterances are to be targets for generation of the learning data, wherein the sorter () is configured to exclude utterances of a dialogue scene that includes utterances similar to utterance of the particular type from the targets for generation of learning data.

Learning data generation device, learning data generation method and non-transitory computer readable recording medium

The learning data generation device () of the present invention comprises: an end-of-talk predict unit () for performing: a first prediction in which it is predicted, based on utterance information on an utterance in the dialog, using the end-of-talk prediction model (), whether the utterance is an end-of-talk utterance of the speaker; and a second prediction in which it is predicted, based on one or more prescribed rules, whether the utterance is an end-of-talk utterance; and a training data generate unit () for generating, when, in the first prediction it is predicted that the utterance is not an end-of-talk utterance and in the second prediction it is predicted that the utterance is an end-of-talk utterance, for the utterance information on the utterance, learning data to which training data indicating that the utterance is an end-of-talk utterance is appended.

A display device for displaying an utterance and information extracted from the utterance, includes an input/output interface configured to display display blocks for a series of dialogue scenes in chronological order of acquisition of utterances, each of the series of dialogue scenes being indicated by dialogue scene data stored in correspondence with utterance data indicating a corresponding one of the utterances, display dialogue scene information within each of the display blocks for the series of dialogue scenes, the dialogue scene information including the corresponding one of the utterances, an utterance type indicating a type of the corresponding one of the utterances, or utterance focus point information of the corresponding one of the utterances, and switch, based on an operation input, between displaying the dialogue scene information and not displaying the dialogue scene information within each of the display blocks for the series of dialogue scenes.

Device and method for supporting creation of reception history, non-transitory computer readable recording medium

The present invention makes it possible to efficiently create an appropriate dialogue history. This device for supporting creation of dialogue history () is provided with: a dialogue utterance focus point information store () which, according to utterance data indicating utterances, stores dialogue scene data indicating dialogue scenes of the utterances, utterance type indicating the types of the utterances, and utterance focus point information of the utterances; and an input/output interface () which, with respect to each of the dialogue scenes indicated by the dialogue scene data stored in the dialogue utterance focus point information store (), causes a display device to display any one or more of utterances, utterance type, and utterance focus point information. Based on an operation input to the input/output interface (), the dialogue utterance focus point information store () adds, modifies, or deletes any one or more of the dialogue scene data, the utterance type, and the utterance focus point information.

To provide a speech recognition results re-ranking technology for re-ranking speech recognition results so as to render speech recognition results suitable for intended use of speech recognition while reducing preparation costs required prior to execution of re-ranking processing of speech recognition results. A speech recognition results re-ranking device includes: a speech recognition unitthat generates a speech recognition result set with recognition score from speech data; and a re-ranking unitthat generates a speech recognition result set with integrated score from the speech recognition result set with recognition score by using a word vector expression database, a cluster center vector expression database, and a normalized knowledge information word DF value database.

Growing community of inventors

Yokosuka, Japan

Setsuo Yamada