-
51.发明申请公开(公告)号：US20180137855A1
公开(公告)日：2018-05-17
申请号：US15598966
申请日：2017-05-18
Applicant: SAMSUNG ELECTRONICS CO., LTD.
Inventor： Sangho LEE , Hyoungmin PARK
CPC classification number: G10L15/1815 , G06F17/273 , G06F17/2735 , G06F17/279 , G10L15/02 , G10L15/063 , G10L15/14 , G10L15/16 , G10L15/19 , G10L15/22 , G10L2015/025
Abstract: A natural language processing method and corresponding apparatus are disclosed. The natural language processing method may include converting words in sentence data, recognized through voice recognition, to corresponding word vectors, and converting characters in the sentence data to corresponding character vectors. The natural language processing method also may include generating a sentence vector based on the word vectors and the character vectors, and determining intent information of the sentence data based on the sentence vector.
-
52.发明申请公开(公告)号：US20180129795A1
公开(公告)日：2018-05-10
申请号：US15678343
申请日：2017-08-16
Applicant: Idefend LTD.
Inventor： Ori KATZ-OZ , Noam ROTEM
CPC classification number: G06F21/32 , G10L15/005 , G10L15/02 , G10L15/1822 , G10L17/22 , G10L17/24 , G10L2015/025 , H04L29/06809 , H04L63/0861 , H04L63/10 , H04W12/06
Abstract: The present invention provides a method for authenticate a user access or action using a computerized device, using audio data inputted by the user, said method implemented by one or more processors operatively coupled to a non-transitory computer readable storage device, on which are stored modules of instruction code that when executed cause the one or more processors to perform: a. at a time preceding a logging attempt, identify and recording user authentic phonetic recording; b. generating selected of words that the user has to verbally repeat; c. recording the user's audio data of saying said selected words; d. phonetically parsing the audio recording of the selected words that was spoken by the user; e. comparing the parsed phonetics of the selected to the user's recorded authenticated phonetic information; and f. assigning a authentication score based on compatibility degree of matching user's phonetic information matched to the authenticated phonetic information.
-
公开(公告)号：US20180124356A1
公开(公告)日：2018-05-03
申请号：US15798428
申请日：2017-10-31
Applicant: Fermax Design & Development, S.L.U.
Inventor： Carlos FERRER ZAERA , Jose Ignacio Garcia Bort , Vicente Albert Perez
CPC classification number: H04N7/147 , G10L13/02 , G10L15/02 , G10L15/22 , G10L25/60 , G10L2015/025 , G10L2015/223 , G10L2015/228 , H04N5/378 , H04N7/186
Abstract: An accessible electronic door entry system that includes an outdoor panel that comprises a capturing microphone, an analog audio interface that digitizes sound, a threshold detector that discriminates the quality of the sound, acoustic models that represent the pronunciation of phonemes, a phoneme generator, contexts that represent the assembly of words and/or phrases and grammatical rules, a recognizer that compares the phonemes, an analyzer of words and/or phrases, a text-to-speech or TTS converter, an analog audio interface that converts the digital signals into analog ones, a communications bus of the electronic door entry system, an electronic door entry system interface that connects the bus and transmits the detected commands to the terminals and establishes the audio and/or video communication, a loudspeaker that plays the audio signals, an agenda, a RAM memory, a Flash memory and a CPU or processor that controls and manages the rest of the elements of said panel.
-
54.发明申请公开(公告)号：US20180114525A1
公开(公告)日：2018-04-26
申请号：US15850106
申请日：2017-12-21
Applicant: INTERACTIVE INTELLIGENCE GROUP, INC.
Inventor： Vivek Tyagi , Aravind Ganapathiraju , Felix Immanuel Wyss
CPC classification number: G10L15/063 , G10L15/144 , G10L2015/025
Abstract: A system and method are presented for acoustic data selection of a particular quality for training the parameters of an acoustic model, such as a Hidden Markov Model and Gaussian Mixture Model, for example, in automatic speech recognition systems in the speech analytics field. A raw acoustic model may be trained using a given speech corpus and maximum likelihood criteria. A series of operations are performed, such as a forced Viterbi-alignment, calculations of likelihood scores, and phoneme recognition, for example, to form a subset corpus of training data. During the process, audio files of a quality that does not meet a criterion, such as poor quality audio files, may be automatically rejected from the corpus. The subset may then be used to train a new acoustic model.
-
公开(公告)号：US20180068661A1
公开(公告)日：2018-03-08
申请号：US15811586
申请日：2017-11-13
Applicant: Promptu Systems Corporation
Inventor： Harry William Printz
CPC classification number: G10L15/22 , G01C21/3608 , G06F3/167 , G06F17/278 , G10L15/02 , G10L15/16 , G10L15/1815 , G10L15/19 , G10L15/32 , G10L2015/025 , G10L2015/228
Abstract: Various embodiments contemplate systems and methods for performing automatic speech recognition (ASR) and natural language understanding (NLU) that enable high accuracy recognition and understanding of freely spoken utterances which may contain proper names and similar entities. The proper name entities may contain or be comprised wholly of words that are not present in the vocabularies of these systems as normally constituted. Recognition of the other words in the utterances in question, e.g. words that are not part of the proper name entities, may occur at regular, high recognition accuracy. Various embodiments provide as output not only accurately transcribed running text of the complete utterance, but also a symbolic representation of the meaning of the input, including appropriate symbolic representations of proper name entities, adequate to allow a computer system to respond appropriately to the spoken request without further analysis of the user's input.
-
Patent Agency Ranking
公开(公告)号：US09898459B2
公开(公告)日：2018-02-20
申请号：US14855346
申请日：2015-09-15
Applicant: VOICEBOX TECHNOLOGIES CORPORATION
Inventor： Min Tang
CPC classification number: G06F17/28 , G06F17/277 , G06F17/2775 , G06F17/2785 , G10L15/1815 , G10L2015/025
Abstract: The invention relates to a system and method for integrating domain information into state transitions of a Finite State Transducer (“FST”) for natural language processing. A system may integrate semantic parsing and information retrieval from an information domain to generate an FST parser that represents the information domain. The FST parser may include a plurality of FST paths, at least one of which may be used to generate a meaning representation from a natural language input. As such, the system may perform domain-based semantic parsing of a natural language input, generating more robust meaning representations using domain information. The system may be applied to a wide range of natural language applications that use natural language input from a user such as, for example, natural language interfaces to computing systems, communication with robots in natural language, personalized digital assistants, question-answer query systems, and/or other natural language processing applications.
-
公开(公告)号：US20180020305A1
公开(公告)日：2018-01-18
申请号：US15209145
申请日：2016-07-13
Applicant: Hand Held Products, Inc.
Inventor： David D. Hardek
CPC classification number: H04R29/004 , G10L15/01 , G10L15/16 , G10L15/26 , G10L2015/025 , G10L2015/0631 , H04R5/033 , H04R2201/107
Abstract: A method for determining a relative position of a microphone may include capturing speech audio from a user's mouth with the microphone so that the microphone outputs an electrical signal indicative of the speech audio; determining an indication of a position of the microphone relative to the user's mouth, which may include providing a plurality of inputs to a computerized discriminative classifier, wherein an input of the plurality of inputs is derived from the electrical signal, and wherein an output from the computerized discriminative classifier is indicative of the position of the microphone relative to the user's mouth.
-
公开(公告)号：US20170329841A1
公开(公告)日：2017-11-16
申请号：US15154000
申请日：2016-05-13
Applicant: Avaya Inc.
Inventor： Wendy J. Holmes , David Skiba
IPC: G06F17/30 , G10L15/02 , G10L15/183 , G06F17/27
CPC classification number: G06F16/3343 , G06F16/3329 , G06F17/2735 , G10L15/02 , G10L25/54 , G10L2015/025
Abstract: A method, system, and phonetic search engine are described that enable phonetic searches to have increased relevancy to the searcher. Specifically, phonetic searches on a database containing phonetically-searchable content can have one or more phonetically-confusable terms included therein, thereby creating search results that more faithfully reflect the search terms used during the phonetic search of the database.
-
公开(公告)号：US09818410B2
公开(公告)日：2017-11-14
申请号：US14983315
申请日：2015-12-29
Applicant: Google Inc.
Inventor： Hasim Sak , Andrew W. Senior
CPC classification number: G10L17/14 , G06N3/0445 , G10L15/02 , G10L15/16 , G10L2015/025
Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media for learning pronunciations from acoustic sequences. One method includes receiving an acoustic sequence, the acoustic sequence representing an utterance, and the acoustic sequence comprising a sequence of multiple frames of acoustic data at each of a plurality of time steps; stacking one or more frames of acoustic data to generate a sequence of modified frames of acoustic data; processing the sequence of modified frames of acoustic data through an acoustic modeling neural network comprising one or more recurrent neural network (RNN) layers and a final CTC output layer to generate a neural network output, wherein processing the sequence of modified frames of acoustic data comprises: subsampling the modified frames of acoustic data; and processing each subsampled modified frame of acoustic data through the acoustic modeling neural network.
-
公开(公告)号：US09818401B2
公开(公告)日：2017-11-14
申请号：US15269924
申请日：2016-09-19
Applicant: Promptu Systems Corporation
Inventor： Harry William Printz
CPC classification number: G10L15/19 , G01C21/3608 , G06F17/278 , G10L15/02 , G10L15/05 , G10L15/1815 , G10L15/22 , G10L15/32 , G10L2015/025 , G10L2015/088 , G10L2015/221 , G10L2015/223
Abstract: Various embodiments contemplate systems and methods for performing automatic speech recognition (ASR) and natural language understanding (NLU) that enable high accuracy recognition and understanding of freely spoken utterances which may contain proper names and similar entities. The proper name entities may contain or be comprised wholly of words that are not present in the vocabularies of these systems as normally constituted. Recognition of the other words in the utterances in question, e.g. words that are not part of the proper name entities, may occur at regular, high recognition accuracy. Various embodiments provide as output not only accurately transcribed running text of the complete utterance, but also a symbolic representation of the meaning of the input, including appropriate symbolic representations of proper name entities, adequate to allow a computer system to respond appropriately to the spoken request without further analysis of the user's input.
-
-
-
-
-
-
-
-
-
Search Results
1111
records
(took 0.034 seconds)
