公开(公告)号：US20210019477A1

公开(公告)日：2021-01-21

申请号：US16988551

申请日：2020-08-07

申请人： Google LLC

发明人： Jakob D. Uszkoreit ,  Mitchell Thomas Stern ,  Jamie Ryan Kiros ,  William Chan

IPC分类号： G06F40/44 ,  G06N3/04 ,  G06N3/08 ,  G06N5/04

摘要： Methods, systems, and apparatus, including computer programs encoded on computer storage media, for generating network outputs using insertion operations.

公开(公告)号：US10885285B2

公开(公告)日：2021-01-05

申请号：US16116833

申请日：2018-08-29

申请人： Google LLC

发明人： Franz Josef Och ,  Jeffrey Dean ,  Thorsten Brants ,  Alexander Mark Franz ,  Jay Ponte ,  Peng Xu ,  Sha-Mayn Teh ,  Jeffrey Chin ,  Ignacio E. Thayer ,  Anton Carver ,  Daniel Rosart ,  John S. Hawkins ,  Karel Driesen

IPC分类号： G06F40/58 ,  G06F40/44 ,  G06F40/47 ,  G06F40/49

摘要： Systems, methods, and apparatus for accessing distributed models in automated machine processing, including using large language models in machine translation, speech recognition and other applications.

公开(公告)号：US10860808B2

公开(公告)日：2020-12-08

申请号：US16267207

申请日：2019-02-04

申请人： Alibaba Group Holding Limited

发明人： Kai Song ,  Weihua Luo ,  Feng Lin

IPC分类号： G06F17/26 ,  G06F40/44 ,  G06F40/30 ,  G06F40/55 ,  G06F40/205

摘要： Implementations herein relate to methods and devices for generating candidate translations and for quantizing text as well as words. A method may include generating, by a computing device, pending candidate translations of text to be translated based on predetermined translation rules. The computing device may generate translation probabilities from the text to be translated to the pending candidate translations based on features having impacts on translation probabilities of the pending candidate translations and a predetermined translation probability prediction model. The computing device may then select a predetermined number of pending candidate translations that have the translation probabilities higher than other pending candidate translations in the pending candidate translations to be the candidate translations of the text to be translated. In implementations, the features having impacts on the translation probabilities may include degrees of semantic similarity between the text to be translated and the candidate translations.

公开(公告)号：US20200380216A1

公开(公告)日：2020-12-03

申请号：US16994347

申请日：2020-08-14

申请人： Amazon Technologies, Inc.

发明人： Hagen Fuerstenau ,  Felix Hieber

IPC分类号： G06F40/44 ,  G06N3/08 ,  G06N20/00 ,  G06F40/51 ,  G06F40/55 ,  G06F40/58

摘要： Based on a candidate set of translations produced by a neural network based machine learning model, a mapping data structure such as a statistical phrase table is generated. The mapping data structure is analyzed to obtain a quality metric of the neural network based model. One or more operations are initiated based on the quality metric.

公开(公告)号：US20200380214A1

公开(公告)日：2020-12-03

申请号：US16998915

申请日：2020-08-20

申请人： Oracle International Corporation

发明人： Boris Galitsky

IPC分类号： G06F40/35 ,  G06F16/31 ,  G06F16/332 ,  G06F40/44 ,  G06F40/51 ,  G06F40/55 ,  G06F40/211 ,  G06K9/62 ,  G06N3/08

摘要： Systems, devices, and methods of the present invention calculate a rhetorical relationship between one or more sentences. In an example, a computer-implemented method accesses a sentence comprising a plurality of fragments. At least one fragment includes a verb and a words. Each word includes a role of the words within the fragment. Each fragment is an elementary discourse unit. The method generates a discourse tree that represents rhetorical relationships between the sentence fragments. The discourse tree includes nodes including nonterminal and terminal nodes, each nonterminal node representing a rhetorical relationship between two of the sentence fragments, and each terminal node of the nodes of the discourse tree is associated with one of the sentence fragments. The method matches each fragment that has a verb to a verb signature, thereby creating communicative discourse tree.

公开(公告)号：US10789431B2

公开(公告)日：2020-09-29

申请号：US16031021

申请日：2018-07-10

申请人： YANDEX EUROPE AG

发明人： Sergey Dmitrievich Gubanov ,  Anton Aleksandrovich Dvorkovich ,  Boris Andreevich Kovarsky ,  Mikhail Alekseevich Nokel ,  Aleksey Anatolievich Noskov ,  Anton Viktorovich Frolov

IPC分类号： G06F40/44 ,  G06F40/58 ,  G06F40/51 ,  G06F40/30 ,  G06F40/42

摘要： A method and system for translating a source sentence in a first language into a target sentence in a second language is disclosed. The method comprises acquiring the source sentence and generating, a first translation hypothesis and a second translation model. A first score value is assigned to the first translation hypothesis, the first score value being representative of a likelihood of the first translation hypothesis is one of a semantically illogical translation. A second score value is assigned to the first translation hypothesis, the second score value being representative of an expected difference in translation quality between the first translation hypothesis and the second translation hypothesis. The target sentence corresponds to: the first translation hypothesis, upon determining that both the first score value and the second score value meet a condition; and the second translation hypothesis, upon determining that the condition is not met.

公开(公告)号：US10769386B2

公开(公告)日：2020-09-08

申请号：US15831395

申请日：2017-12-05

申请人： SAP SE

发明人： Michail Vasiltschenko ,  Benjamin Schork ,  Mark Childress ,  Annika Berger ,  Yvonne Miklosch ,  Thomas Alcantara Kliche ,  Hanne Smaadahl ,  Ying Zou

IPC分类号： G06F40/44 ,  G06F16/338 ,  G06F16/33 ,  G06F40/20 ,  G06F40/131 ,  G06F40/157 ,  G06F40/242 ,  G06F40/247 ,  G06F40/263 ,  G06F40/40 ,  G06F40/58 ,  G06F40/47 ,  G06F40/51

摘要： An input terminology is received in a source language to determine a target language proposal. In a proposal engine when a configuration profile is available the following steps are performed. A set of target language equivalents are determined by applying a user-defined set of approaches to the input terminology. Statistics is computed corresponding to the set of target language equivalents. A target language proposal from the set of target language equivalents is determined based on the computed statistics. The target language proposal along with the computed statistics is displayed in a graphical user interface.

公开(公告)号：US10755054B1

公开(公告)日：2020-08-25

申请号：US16010168

申请日：2018-06-15

申请人： Facebook, Inc.

发明人： Alexander Waibel ,  Matthias Paulik

IPC分类号： G06F17/27 ,  G06F17/28 ,  G06F40/58 ,  G10L15/00 ,  G06F40/44 ,  G06F40/263 ,  G06F40/30

摘要： An iterative language translation system includes multiple communicatively connected statistical speech translation systems. The system includes an automatic speech recognition component adapted to recognize spoken language in a source language and to create a source language hypothesis. A machine translation component is adapted to translate the source language hypothesis into a target language. The system also includes a second automatic speech recognition component and second machine translation component. The translation results are used to adapt the automatic speech recognition components and the language hypotheses are used to adapt the machine translation components.

公开(公告)号：US20200175233A1

公开(公告)日：2020-06-04

申请号：US16745947

申请日：2020-01-17

申请人： Samsung Electronics Co., Ltd.

发明人： Jihyun LEE ,  Hoshik LEE

IPC分类号： G06F40/47 ,  G06F40/30 ,  G06F40/44 ,  G06F40/56 ,  G06F40/242 ,  G06F40/247

摘要： A sentence generating method includes: generating a corresponding word set of a source word set generated based on a source sentence; generating words by performing decoding based on feature vectors generated through encoding of the source sentence; adjusting a probability of at least one of the generated words based either one or both of the source word set and the corresponding word set; and selecting character strings from different character strings including each of the generated words based on the adjusted probability and the probability as unadjusted.

公开(公告)号：US10664666B2

公开(公告)日：2020-05-26

申请号：US15854058

申请日：2017-12-26

申请人： BEIJING BAIDU NETCOM SCIENCE AND TECHNOLOGY CO., LTD.

发明人： Shiqi Zhao

IPC分类号： G06F40/58 ,  G06N20/00 ,  G06F40/44 ,  G06F40/45 ,  G06F40/47 ,  G06F40/279 ,  G06F40/53 ,  G06F40/49 ,  G06F40/55 ,  G06F40/51

摘要： Provided are a language conversion method and device based on artificial intelligence, and a terminal. The language conversion method includes: receiving a language conversion request, wherein the language conversion request comprises a sentence to be converted, and the sentence to be converted comprises an original language word and a target language word; determining a plurality of candidate words each corresponding to the original language word according to a preset rule; determining a score of each of the plurality of candidate words when used in the sentence to be converted; and determining a target word corresponding to the original language word according to the score of each of the plurality of candidate words.