-
1.发明申请公开(公告)号：US20180366120A1
公开(公告)日：2018-12-20
申请号：US15989603
申请日：2018-05-25
Inventor： TAKASHI USHIO , KATSUYOSHI YAMAGAMI , MITSURU ENDO
Abstract: A speech continuation determination method includes acquiring dialogue data including a system sentence spoken to a user at a first time, a user sentence spoken by the user at a second time following the system sentence, and system intention indicating intention of the system sentence; estimating a predicted response delay amount indicating a wait time for a response to the user sentence by applying the dialogue data to a model obtained by machine learning; acquiring user status information indicating the status of the user; and determining whether a speech sentence by the user continues following the user sentence in accordance with the user status information in the wait time indicated by the predicted response delay amount.
-
2.发明申请公开(公告)号：US20180158459A1
公开(公告)日：2018-06-07
申请号：US15821747
申请日：2017-11-22
Inventor： KATSUYOSHI YAMAGAMI , MITSURU ENDO , TAKASHI USHIO
Abstract: An information processing method for information stored in a storage includes: holding a dialog history of a dialog including a question to a user and a reply from the user to the question, determining whether a manner in which a third reply indicating neither a first reply nor a second reply appears in a reply history of the reply included in the held dialog history satisfies a predetermined condition, the first reply indicating an affirmative in response to the question, the second reply indicating a negative in response to the question; and performing presentation regarding the information stored in the storage if the manner is determined to satisfy the predetermined condition.
-
3.发明申请公开(公告)号：US20180150553A1
公开(公告)日：2018-05-31
申请号：US15805915
申请日：2017-11-07
Inventor： MITSURU ENDO , KATSUYOSHI YAMAGAMI , TAKASHI USHIO
IPC: G06F17/30
CPC classification number: G06F16/3344 , G06F16/735 , G06F16/90332
Abstract: There is provided an information processing method for performing, through interaction with a user, narrowing down regarding information that the user desires to search for. The method includes: outputting a first question, which is an open question, about a target of search; obtaining a first answer to the first question, the first answer being input by the user; if a determination is made indicating that the first answer does not satisfy a first condition and that a word or phrase corresponding to a word or phrase included in the first answer is not included in a database, outputting a second question, which is an open question, for requesting an explanation about the word or phrase included in the first answer and not included in the database; and if a determination is made indicating that the first answer satisfies the first condition, outputting a closed question about the target of search.
-
4.发明申请公开(公告)号：US20170270909A1
公开(公告)日：2017-09-21
申请号：US15448688
申请日：2017-03-03
Inventor： MAYU YOKOYA , KATSUYOSHI YAMAGAMI
Abstract: A method includes determining a content of a speech of a user on the basis of text data including a word string acquired as a recognition result of the speech, selecting a correction method suitable for the determined content of the speech from among multiple correction methods for correcting false recognition contained in the text data, and correcting the false recognition contained in the text data using the selected correction method.
-
公开(公告)号：US20170084269A1
公开(公告)日：2017-03-23
申请号：US15262785
申请日：2016-09-12
Inventor： HONGJIE SHI , TAKASHI USHIO , MITSURU ENDO , KATSUYOSHI YAMAGAMI
CPC classification number: G10L15/16 , G06F17/2785 , G06N3/0454 , G10L15/063 , G10L2015/0635
Abstract: A subject estimation system includes a convolutional neural network to estimate a subject label of a dialog. The convolution neural network includes: one or more topic-dependent convolutional layers and one topic-independent convolutional layer, each of the one or more topic-dependent convolutional layers performing, on an input of a word-string vector sequence corresponding to dialog text transcribed from a dialog, a convolution operation dependent on a topic, and the topic-independent convolutional layer performing, on the input of the word-string vector sequence, a convolution operation not dependent on the topic; a pooling layer performing pooling process on outputs of the convolutional layer; and a fully connected layer performing full connection process on outputs of the pooling layer.
-
Patent Agency Ranking
公开(公告)号：US20180046614A1
公开(公告)日：2018-02-15
申请号：US15665951
申请日：2017-08-01
Inventor： TAKASHI USHIO , HONGJIE SHI , MITSURU ENDO , KATSUYOSHI YAMAGAMI
IPC: G06F17/27 , G06N99/00 , G06N3/04 , G10L21/0272
CPC classification number: G06F17/2755 , G06F17/2785 , G06F17/279 , G06N3/0445 , G06N3/0454 , G06N20/00 , G10L15/1822 , G10L17/04 , G10L21/0272
Abstract: A dialogue act estimation method includes acquiring learning data including a first sentence to be estimated in the form of text data of a first uttered sentence uttered at a first time point, a second sentence which is text data of a second uttered sentence uttered, at a time point before the first time point, successively after the first uttered sentence, act information indicating an act associated to the first sentence, property information indicating a property information associated to the first sentence, and dialogue act information indicating a dialogue act in the form of a combination of an act and a property associated to the first sentence, making a particular model learn three or more tasks at the same time using the learning data, and storing a result of the learning as learning result information in a memory.
-
公开(公告)号：US20170365252A1
公开(公告)日：2017-12-21
申请号：US15618383
申请日：2017-06-09
Inventor： TAKASHI USHIO , KATSUYOSHI YAMAGAMI
CPC classification number: G10L15/063 , G06F17/2785 , G06N3/0445 , G10L15/16 , G10L15/1815
Abstract: A meaning generation method, in a meaning generation apparatus, includes acquiring meaning training data including text data of a sentence that can be an utterance sentence and meaning information indicating a meaning of the sentence and associated with the text data of the sentence, acquiring restatement training data including the text data of the sentence and text data of a restatement sentence of the sentence, and learning association between the utterance sentence and the meaning information and the restatement sentence. The learning includes learning of a degree of importance of a word included in the utterance sentence, and the learning is performed by applying the meaning training data and the restatement training data to a common model, and storing a result of the learning as learning result information.
-
公开(公告)号：US20180046619A1
公开(公告)日：2018-02-15
申请号：US15662628
申请日：2017-07-28
Inventor： HONGJIE SHI , TAKASHI USHIO , MITSURU ENDO , KATSUYOSHI YAMAGAMI
Abstract: A method for controlling identification includes obtaining first text, which is text in a first language, obtaining second text, which is text in a second language obtained by translating the first text into the second language, obtaining correct labels, which indicate content of the first text, inputting the first text and the second text to an identification model common to the first and second languages, and updating the common identification model such that labels identified by the common identification model from the first text and the second text match the correct labels.
-
公开(公告)号：US20180005333A1
公开(公告)日：2018-01-04
申请号：US15627350
申请日：2017-06-19
Inventor： MITSURU ENDO , KATSUYOSHI YAMAGAMI
IPC: G06Q50/22 , G06Q50/16 , G06F3/0489 , H04R1/02 , G06F3/0488
CPC classification number: G06Q50/22 , G06F3/0488 , G06F3/0489 , G06Q50/16 , G16H10/20 , G16H50/20 , H04R1/021
Abstract: An information processing method in an interactive apparatus that questions a user's symptom through interaction with a user includes: outputting first question information of a first type concerning the user's symptom to a display or a speaker connected to the interactive apparatus, receiving first answer information indicating an answer to the question in the first question information from a keyboard, a touch panel, or a microphone connected to the interactive apparatus, outputting second question information of the first type concerning the user's symptom to the display or the speaker when it is determined that the answer in the first answer information does not include a word in a negative expression, and outputting third question information of a second type concerning the user's symptom to the display or the speaker when it is determined that the answer in the first answer information includes the word in the negative expression.
-
公开(公告)号：US20180005126A1
公开(公告)日：2018-01-04
申请号：US15626183
申请日：2017-06-19
Inventor： KATSUYOSHI YAMAGAMI , MITSURU ENDO
CPC classification number: G06N5/045 , G06F16/9027 , G06N5/003 , G06N20/00 , G06Q30/02
Abstract: A decision tree generating apparatus includes an information gain calculator and a decision tree generator. When a classification target data set including a plurality of pieces of classification target data respectively having different attributes with attribute values assigned thereto is segmented into subsets in a form of a decision tree, the information gain calculator calculates an amount of entropy reduction on each attribute, and calculates an information gain, based on the amount of reduction in the entropy and reliability of a user's answer responsive to an inquiry asking about the attribute. The decision tree generator successively determines an attribute having a maximum information gain to be a node of the decision tree by recursively iterating the segmentation of the pre-segmentation data set, and generates the decision tree that is to be used to determine an order of the inquiries.
-
-
-
-
-
-
-
-
-
Search Results
11
records
(took 0.019 seconds)
