WO2012002841A1 - Procédé de reconnaissance du message d'une personne - Google Patents

Procédé de reconnaissance du message d'une personne Download PDF

Info

Publication number
WO2012002841A1
WO2012002841A1 PCT/RU2011/000421 RU2011000421W WO2012002841A1 WO 2012002841 A1 WO2012002841 A1 WO 2012002841A1 RU 2011000421 W RU2011000421 W RU 2011000421W WO 2012002841 A1 WO2012002841 A1 WO 2012002841A1
Authority
WO
WIPO (PCT)
Prior art keywords
person
ias
individual
speech
algorithm
Prior art date
Application number
PCT/RU2011/000421
Other languages
English (en)
Russian (ru)
Inventor
Владимир Витальевич МИРОШНИЧЕНКО
Виталий Евгеньевич ПИЛКИН
Original Assignee
Miroshnichenko Vladimir Vitalievich
Pilkin Vitaly Evgenievich
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Miroshnichenko Vladimir Vitalievich, Pilkin Vitaly Evgenievich filed Critical Miroshnichenko Vladimir Vitalievich
Publication of WO2012002841A1 publication Critical patent/WO2012002841A1/fr

Links

Classifications

    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
    • G10L15/00—Speech recognition
    • G10L15/24—Speech recognition using non-acoustical features

Definitions

  • the invention relates to a method for recognizing human speech and / or non-speech messages and can be used in communication between people, including between people speaking different languages.
  • the objective of the proposed technical solution is to create a method for recognizing human speech and / or non-speech messages by using an individual correspondence algorithm, including: a) an individual human speech algorithm and / or b) an individual human speech reduction algorithm and / or c) an individual correspondence algorithm movements and / or gestures and / or actions of a person to a voice message and / or g) individual human facial expressions algorithm or any combination consisting in whole or in part of the above options.
  • an electronic message recognition program can recognize a speech and non-speech message of any person regardless of their individual characteristics of pronunciation of voice messages and / or non-voice messages.
  • An electronic program with a message recognition function (hereinafter referred to as EPRS) is an electronic program with a function for recognizing speech and / or gesture and / or other non-speech messages of a person, which, when recognizing a person’s message, can be connected to at least one directed to the specified a person with a video camera and / or with a device by and / or through which control the specified or specified video cameras.
  • An individual speech algorithm is one or more files recorded or recorded on at least one electronic device using at least one electronic device in digital format, in which or in which: 1) the difference between a person’s pronunciation is described letters and / or combinations of letters and / or words and / or phrases and the pronunciation of the indicated individual letters and / or combinations of letters and / or words and / or phrases recorded in the memory of at least one indicated electronic a program that (difference) is taken into account by EPRS when recognizing the speech of a specified person, and / or 2) a dictionary of correspondences of a single letter and / or combination of letters and / or words and / or phrases of a single letter and / or a combination of letters and / or a word and / or phrase recorded in the memory of at least one of the indicated electronic programs and / or 3) includes information on the individual characteristics of human speech and / or information related to the individual characteristics of human speech, which can be used by EPRS in recognition speech of a pointed person. It is possible to completely or partially change and and
  • An individual speech abbreviation algorithm is one or more files recorded using at least one electronic device through at least one electronic program in digital format, in which or which includes a list of word abbreviations and / or a list of word abbreviations.
  • EPRS using the individual algorithm of speech contractions of the specified person, translates these abbreviations into words or phrases assigned to them. It is possible to completely or partially change and / or delete and / or update the specified list and / or its part and / or add the indicated correspondences to the specified list and / or its part.
  • An individual algorithm for matching movements and / or gestures and / or actions of a person with a voice message is one or more files recorded using at least one electronic device through at least one electronic program in digital format, in which or in which a list of matches is included one movement and / or at least one gesture and / or at least one action of the specified person to at least one word or at least one phrase.
  • the EPRS using an individual algorithm for matching movements and / or gestures and / or actions of the specified person with the speech message, translates the movements and / or gestures and / or actions into the corresponding words or phrases. It is possible to completely or partially change and / or delete and / or update the specified list and / or its part and / or add the indicated correspondences to the specified list and / or its part.
  • An individual facial expression algorithm is recorded using at least one electronic device using at least one electronic program in digital format, at least one or more files, which includes or includes a list of facial expressions on a person’s face pronunciation of at least one letter or at least one word or at least one phrase.
  • EPRS using an individual facial expression algorithm, translates movements and / or changes in facial expressions on a person’s face into the corresponding letters or words or phrases. It is possible to completely or partially modify and / or delete and / or update the specified list and / or its part and / or add the indicated correspondences to the specified list and / or its part.
  • An electronic translator program is an electronic program that translates a person’s recognized EPRS message into another at least one speech language and / or another at least one sign language and / or translates a recognized EPRS message from sign language to speech language (if communication was carried out in sign language) and / or translates from a speech language into a sign language.
  • IAS individual correspondence algorithm
  • US includes: a) an individual speech algorithm of a specified person and / or b) an individual speech algorithm abbreviations of a specified person and / or c) an individual algorithm for matching movements and / or gestures and / or actions of a specified person with a speech message and / or d) an individual algorithm for the facial expressions of a specified person or any combination consisting entirely or partially of the options specified in subparagraphs a), b), c), d).
  • a human IAS can be information in electronic form or information in electronic form and at least one electronic program.
  • EPRS message recognition function
  • EPRS can be configured to interact with at least one electronic device with or without a display and / or with at least one electronic translator and / or at least one video camera and / or the Internet via wireless and / or wired communication. 5.
  • a person’s IAS can be fully or partially placed in an electronic device and / or in the memory of an external electronic device to which the indicated electronic device is connected, and / or on the Internet.
  • Access to IAS can be carried out by means of wireless and / or wire communication.
  • a person’s IAS can have at least one individual code.
  • a person’s IAS may include: a) information about the language spoken by the specified person, or which sign language the person is communicating with, or the dialect of the spoken language the specified person speaks and / or b) at least one code (login , password) access to the person’s IAS and / or part of the person’s IAS and / or c) at least one multimedia file and / or at least one computer program and / or d) information about the specified person and / or information that is associated with the specified by man.
  • the IAS may completely or partially change and / or from the IAS may completely or partially delete and / or may add: a) information and / or an indication of the speech and / or sign language and / or dialect of a person who has IAS, and / or b) at least one access code to the IAS and / or c) at least one multimedia file and / or at least one computer program and / or d) information about the specified person and / or information related to specified person.
  • the invention is practicable because it uses hardware and software known from the prior art.
  • Example 1 The individual characteristics of human speech, an individual human speech algorithm and an individual human speech reduction algorithm are created by: a) repeating the letters, words, phrases, any combinations, text or texts spoken by the electronic device into the microphone by the person creating his IAS ( hereinafter - EU), or an external electronic device to which the EU is connected, or b) reading aloud by the person creating his IAS, letters, words, phrases, any combinations thereof, text or texts provided by the electronic control unit or an external electronic device to which the electronic control device is connected, in the form of a text message; and b) the compliance is stored electronically and taken into account when using IAS.
  • - EU the person creating his IAS
  • Example 2 The individual characteristics of human speech, the individual human speech algorithm and the individual human speech reduction algorithm are created in the following way: letters, words, phrases, any combinations of them spoken into the microphone by the person creating their IAS, are checked by the electronic control unit or an external electronic device to which it is connected EU, according to their correspondence to letters, words, phrases, any combinations thereof, recorded in the memory of the EU or in the memory of an external electronic device; if the control unit or an external electronic device to which the control unit is connected voiced and / or written on the display a letter and / or word and / or phrase spoken incorrectly by the indicated person, the specified person corrects the allowed control unit or the external electronic device to which the control unit is connected, an error by repeating a letter, a word, a phrase, any combinations thereof that were incorrectly determined by the control unit or an external electronic device to which the control unit is connected, for their compliance and / or by correcting the control unit on the display or on the display the electronic device to which the electronic control device is connected, incorrectly detected electronic control devices or an external
  • Example 3 An individual algorithm for matching a person’s movements and / or gestures with a voice message and an individual person’s facial expressions algorithm are created by video recording on the electronic control unit or on an external electronic device to which the electronic device is connected, using at least one video camera to perform at least one movement and / or performing at least one gesture and / or changing at least one facial expression of his face, which correspond, in the opinion of the indicated person, to the letter and / or word and / or phrase, voiced or voiced PP or PP presentation or presentation in the form of text.
  • the present invention can be used to recognize speech and / or non-speech messages of a person and can be used in communication between people, including between people speaking different languages.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computational Linguistics (AREA)
  • Health & Medical Sciences (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Human Computer Interaction (AREA)
  • Physics & Mathematics (AREA)
  • Acoustics & Sound (AREA)
  • Multimedia (AREA)
  • Telephone Function (AREA)
  • Information Transfer Between Computers (AREA)

Abstract

L'invention concerne un procédé de reconnaissance de messages vocaux et/ou non vocaux d'une personne, qui utilise un algorithme individuel de mise en correspondance comprenant : a) un algorithme individuel de la voix de la personne et/ou b) un algorithme individuel des abréviations vocales de la personne et/ou c) un algorithme individuel de mise en correspondance de mouvements et/ou de gestes et/ou d'actions de la personne par rapport au message vocal et/ou d) un algorithme individuel de mimiques de la personne, ou une quelconque combinaison comprenant entièrement ou partiellement les variantes susmentionnées. L'application de la présente invention permet d'obtenir des avantages importants qui font que le programme électronique de reconnaissance de message permet de reconnaître des messages vocaux ou non vocaux d'une quelconque personne indépendamment de ses particularités individuelles de prononciation de message vocaux et/ou de réalisation de messages non vocaux.
PCT/RU2011/000421 2010-06-29 2011-06-16 Procédé de reconnaissance du message d'une personne WO2012002841A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
RU2010126303/08A RU2010126303A (ru) 2010-06-29 2010-06-29 Распознавание сообщений человека
RU2010126303 2010-06-29

Publications (1)

Publication Number Publication Date
WO2012002841A1 true WO2012002841A1 (fr) 2012-01-05

Family

ID=45402331

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/RU2011/000421 WO2012002841A1 (fr) 2010-06-29 2011-06-16 Procédé de reconnaissance du message d'une personne

Country Status (2)

Country Link
RU (1) RU2010126303A (fr)
WO (1) WO2012002841A1 (fr)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
RU2158485C1 (ru) * 2000-01-24 2000-10-27 Общество с ограниченной ответственностью "Ти Би Кей Интернэшнл" Способ проверки права доступа абонента к системе коллективного пользования
WO2007117814A2 (fr) * 2006-03-29 2007-10-18 Motorola, Inc. Perturbation de signaux vocaux à des fins de reconnaissance vocale
RU80603U1 (ru) * 2008-09-19 2009-02-10 Юрий Константинович Низиенко Электронная приемопередающая система с функцией синхронного перевода устной речи с одного языка на другой
WO2009042579A1 (fr) * 2007-09-24 2009-04-02 Gesturetek, Inc. Interface optimisée pour des communications de voix et de vidéo
RU2389067C2 (ru) * 2003-10-24 2010-05-10 Майкрософт Корпорейшн Системы и способы для проецирования содержимого с компьютерных устройств

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
RU2158485C1 (ru) * 2000-01-24 2000-10-27 Общество с ограниченной ответственностью "Ти Би Кей Интернэшнл" Способ проверки права доступа абонента к системе коллективного пользования
RU2389067C2 (ru) * 2003-10-24 2010-05-10 Майкрософт Корпорейшн Системы и способы для проецирования содержимого с компьютерных устройств
WO2007117814A2 (fr) * 2006-03-29 2007-10-18 Motorola, Inc. Perturbation de signaux vocaux à des fins de reconnaissance vocale
WO2009042579A1 (fr) * 2007-09-24 2009-04-02 Gesturetek, Inc. Interface optimisée pour des communications de voix et de vidéo
RU80603U1 (ru) * 2008-09-19 2009-02-10 Юрий Константинович Низиенко Электронная приемопередающая система с функцией синхронного перевода устной речи с одного языка на другой

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
"Kontseptualnyi LED-telefon-bez-displeya, noyabrya s LED-interfeicom.", MOBBIT: YOUR MOBILITY, 23 January 2008 (2008-01-23), Retrieved from the Internet <URL:http://mobbit.info/item/2008/1/23/konceptyal-nyi-led-telefon-bez-displeya-no-s-led-interft> [retrieved on 20110831] *
VYACHESLAV DOROT ET AL., TOLKOVY SLOVAR SOVREMENNOY KOMPYUTERNOI LEKSIKI, 3-3 IZDANIE, PERERABOTANNOE I DOPOLNENNOE., 2004, ST.PETERSBURG, «BKHV-PETERBURG», pages 221, 491 *

Also Published As

Publication number Publication date
RU2010126303A (ru) 2012-01-10

Similar Documents

Publication Publication Date Title
WO2019165748A1 (fr) Procédé et appareil de traduction vocale
JP5405672B2 (ja) 外国語学習装置及び対話システム
US11145222B2 (en) Language learning system, language learning support server, and computer program product
JP2022527970A (ja) 音声合成方法、デバイス、およびコンピュータ可読ストレージ媒体
KR101183310B1 (ko) 일반적인 철자 기억용 코드
EP1251490A1 (fr) Modèle phonetique compact pour la reconnaissance des langues arabes
CN109461436B (zh) 一种语音识别发音错误的纠正方法及系统
JP6747434B2 (ja) 情報処理装置、情報処理方法、およびプログラム
JP2024508033A (ja) 対話中のテキスト－音声の瞬時学習
KR20100092541A (ko) 중국어 학습장치 및 방법
KR20140071070A (ko) 음소기호를 이용한 외국어 발음 학습방법 및 학습장치
CN104200807B (zh) 一种erp语音控制方法
JP2002244842A (ja) 音声通訳システム及び音声通訳プログラム
KR20150103809A (ko) 유사발음 학습 방법 및 장치
JP5818753B2 (ja) 音声対話システム及び音声対話方法
WO2012002841A1 (fr) Procédé de reconnaissance du message d&#39;une personne
CN107203539B (zh) 复数字词学习机的语音评测装置及其评测与连续语音图像化方法
ES2965480T3 (es) Procesamiento y evaluación de señales del habla
US9437190B2 (en) Speech recognition apparatus for recognizing user&#39;s utterance
JP2022525341A (ja) 聴覚障害者のための触覚的および視覚的意思疎通システム
JP6911696B2 (ja) 修正制御装置、修正制御方法及び修正制御プログラム
JP2005128130A (ja) 音声認識装置、音声認識方法及びプログラム
JP2002244841A (ja) 音声表示システム及び音声表示プログラム
KR20140068292A (ko) 말소리 유창성 향상을 위한 훈련 학습 시스템
AU2020103587A4 (en) A system and a method for cross-linguistic automatic speech recognition

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 11801211

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 11801211

Country of ref document: EP

Kind code of ref document: A1