专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

1. 发明申请

WO2023085506A1 개체명 교정 시스템 및 이를 훈련시키는 방법 审中-公开
公开(公告)号：WO2023085506A1
公开(公告)日：2023-05-19
申请号：PCT/KR2021/018735
申请日：2021-12-10
申请人： 주식회사 솔트룩스
发明人： 이경일 , 박준모
IPC分类号： G06F40/295 , G06F40/232 , G06N3/08 , G06N3/04
摘要： 개체명 교정 시스템이 제공된다. 개체명 교정 시스템은 원문으로부터 원문 개체명들을 획득하고 요약문으로부터 요약문 개체명들을 획득하도록 구성된 개채명 추출 컴포넌트, 상기 원문 개체명들에 대응하는 원문 문장들로부터 상기 원문 개체명들의 원문 개체명 벡터들을 획득하고 상기 요약문 개체명들에 대응하는 요약문 문장들로부터 상기 요약문 개체명들의 요약문 개체명 벡터들을 획득하도록 구성된 언어 컴포넌트, 상기 요약문 개체명 벡터들 중 타겟 요약문 개체명 벡터 및 상기 원문 개체명 벡터들로부터 상기 타겟 요약문 개체명 벡터에 대응하는 타겟 요약문 개체명이 상기 원문에 비추어 옳은지 여부를 판정하도록 구성된 판정 컴포넌트, 및 상기 타겟 요약문 개체명이 옳지 않은 경우 상기 타겟 요약문 개체명을 상기 원문 개체명들 중 하나 또는 상기 원문 개체명들과 다른 새로운 개체명으로 변경하도록 구성된 교정 컴포넌트를 포함하고, 상기 판정 컴포넌트 및 상기 교정 컴포넌트는, 상기 원문 개체명 벡터들 및 상기 타겟 요약문 개체명 벡터로부터 인코더 은닉 상태들, 디코더 은닉 상태, 어텐션 분포, 및 문맥 벡터를 획득하도록 구성된 공통 서브컴포넌트를 공유할 수 있다.

2. 发明申请

WO2022256026A1 SYSTEMS AND METHODS FOR GENERATING PHONETIC SPELLING VARIATIONS 审中-公开
公开(公告)号：WO2022256026A1
公开(公告)日：2022-12-08
申请号：PCT/US2021/043739
申请日：2021-07-29
申请人： GOOGLE LLC
发明人： GUPTA, Abhirut , RAGHUVEER, Aravindan , SHARMA, Abhay , RAUT, Nitin , KUMAR, Manish
IPC分类号： G06F40/232 , G06F40/44 , G10L15/26
摘要： Systems and methods for generating phonetic spelling variations of a given word based on locale-specific pronunciations. A phoneme-letter density model may be configured to identify a phoneme sequence corresponding to an input word, and to identify all character sequences that may correspond to an input phoneme sequence and their respective probabilities. The phoneme-phoneme error model may be configured to identify locale-specific alternative phoneme sequences that may correspond to a given phoneme sequence, and their respective probabilities. Using these two models, a processing system may be configured to generate, for a given input word, a list of alternative character sequences that may correspond to the input word based on locale-specific pronunciations, and/or a probability distribution representing how likely each alternative character sequence is to correspond to the input word.

3. 发明申请

WO2022146861A1 PUNCTUATION AND CAPITALIZATION OF SPEECH RECOGNITION TRANSCRIPTS 审中-公开
公开(公告)号：WO2022146861A1
公开(公告)日：2022-07-07
申请号：PCT/US2021/065040
申请日：2021-12-23
申请人： GENESYS TELECOMMUNICATIONS LABORATORIES, INC.
发明人： FAIZAKOF, Avraham , MAZZA, Arnon , HAIKIN, Lev , ORBACH, Eyal
IPC分类号： G06F40/232 , G06F40/284 , G06N20/00 , G10L15/00
摘要： A method comprising: receiving a first text corpus comprising punctuated and capitalized text; annotating words in said first text corpus with a set of labels indicating a punctuation and a capitalization of each word; at an initial training stage, training a machine learning model on a first training set comprising: (i) said annotated words in said first text corpus, and (ii) said labels; receiving a second text corpus representing conversational speech; annotating words in said second text corpus with said set of labels; at a re-training stage, re-training said machine learning model on a second training set comprising: (iii) said annotated words in said second text corpus, and (iv) said labels; and at an inference stage, applying said trained machine learning model to a target set of words representing conversational speech, to predict a punctuation and capitalization of each word in said target set.

4. 发明申请

WO2022141844A1 文本纠错方法、装置、设备及可读存储介质审中-公开
公开(公告)号：WO2022141844A1
公开(公告)日：2022-07-07
申请号：PCT/CN2021/083277
申请日：2021-03-26
申请人：平安科技（深圳）有限公司
发明人：邓悦 , 郑立颖 , 徐亮
IPC分类号： G06F40/232
摘要：一种文本纠错方法、装置、设备及可读存储介质，涉及人工智能技术领域，该方法包括步骤：获取待纠错文本中的待纠错句子和上文句子，其中，所述上文句子与所述待纠错句子相邻（S110）；基于所述待纠错句子、所述上文句子和预设注意力机制为所述待纠错句子进行评分，得到待纠错注意力评分和上文注意力评分（S120）；叠加所述待纠错注意力评分和所述上文注意力评分，得到纠错评分（S130）；基于所述纠错评分对所述待纠错句子进行纠错，得到纠错后句子（S140）。实现了结合待纠错句子的信息和其上文句子的信息以综合考虑对待纠错句子进行纠错的方案，使得通过对待纠错文本进行纠错后，得到的纠错后的文本的准确率得到提高。

5. 发明申请

WO2021246570A1 컨텐츠를 수정하기 위한 전자 장치 및 방법 审中-公开
公开(公告)号：WO2021246570A1
公开(公告)日：2021-12-09
申请号：PCT/KR2020/010257
申请日：2020-08-04
申请人： 삼성전자 주식회사
发明人： 젤레즈니아코브드미트로 , 야코브츄크올렉 , 올리닉다리나 , 자이체브빅토르 , 야키신예브헨니 , 체르네하아나스타시아
IPC分类号： G06F40/166 , G06F40/232 , G06F40/44 , G06F3/0488 , G06N3/08
摘要： 본 개시는 인공지능 학습 모델을 이용하여 컨텐츠에서 에러가 발생한 위치를 자동적으로 검출하고, 에러를 정정하는 전자 장치 및 방법에 관한 것이다. 구체적으로, 전자 장치는 메모리 및 적어도 하나의 프로세서를 포함할 수 있다. 적어도 하나의 프로세서는 제1 컨텐츠 및 제2 컨텐츠를 순차적으로 수신함으로써, 각각 제1 최종 결과물 및 제2 최종 결과물을 생성할 수 있다. 적어도 하나의 프로세서는 제1 컨텐츠를 인식함으로써 생성된 중간 결과물을 메모리에 저장하며, 제2 최종 결과물 및 중간 결과물에 기초하여, 제1 최종 결과물에 포함된 에러의 위치를 검출함으로써, 제1 최종 결과물을 수정할 수 있다.

6. 发明申请

WO2020263411A1 SYSTEM AND METHOD FOR COOPERATIVE TEXT RECOMMENDATION ACCEPTANCE IN A USER INTERFACE 审中-公开
公开(公告)号：WO2020263411A1
公开(公告)日：2020-12-30
申请号：PCT/US2020/031245
申请日：2020-05-04
申请人： MICROSOFT TECHNOLOGY LICENSING, LLC
发明人： MANNBY, Claes-Fredrik U. , MCGLYNN, Matthew , WU, Yifan
IPC分类号： G06F3/023 , G06F3/0488 , G06F40/274 , G06F40/232
摘要： Methods for cooperative text recommendation acceptance of completion options in a user interface are performed by systems and devices. A user provides inputs via a user interface (UI) that are stored in an input buffer. As a portion of a first input is received, completion options for some part of the first input are determined based on statistical probabilities and the portion. A completion option is selected and displayed via the UI as completing the first input in a differentiated manner from the user-entered input. The user then either generates an acceptance command for the completion option or continues providing the first input and the UI adapts the remaining completion option portion. Acceptance commands are accepted as space characters or as alphanumeric characters representing additional input that follows the first input and the completion option. Statistical likelihoods are used to account for typographical errors and misspellings in user inputs.

7. 发明申请

WO2023064043A1 PERSONALIZED TEXT PROOFING USING DISTRIBUTED SERVICES 审中-公开
公开(公告)号：WO2023064043A1
公开(公告)日：2023-04-20
申请号：PCT/US2022/041434
申请日：2022-08-25
申请人： MICROSOFT TECHNOLOGY LICENSING, LLC.
发明人： COGLEY, James Aidan , CADONI, Enrico , LAIRD, Colin , GUPTA, Shashank Shekhar , GAUTHIER, Olivier
IPC分类号： G06F40/166 , G06F40/253 , G06F40/232 , G06F8/33 , G06F16/93 , G06N20/00
摘要： Systems and methods provide personalized text proofing. A user model that is used to personalize generic critiques for text proofing a document is generated based on user signals indicating past user actions. During runtime of an application used to create the document, the user model is accessed and locally cached. User inputs comprising typed components used to create the document are received and a set of one or more generic critiques for the user inputs is accessed from an editor system. The user model is applied to the set which may modify a generic critique of the set. Modifying the generic critique can cause the generic critique to be automatically applied or suppressed at the client device. The set including the modified generic critique is transmitted to a user device, whereby the user device applies the set to the document including automatically applying or suppressing the modified generic critique.

8. 发明申请

WO2023035525A1 语音识别纠错方法、系统、装置及存储介质审中-公开
公开(公告)号：WO2023035525A1
公开(公告)日：2023-03-16
申请号：PCT/CN2022/071074
申请日：2022-01-10
申请人：平安科技（深圳）有限公司
发明人：庄子扬 , 魏韬 , 马骏 , 王少军 , 肖京
IPC分类号： G06F40/232 , G06F40/216 , G10L15/26
摘要：本申请公开了一种语音识别纠错方法、系统、装置及存储介质，涉及人工智能技术，该方法包括：对待检测语音进行语音识别，获取到待检测文本和对应的待检测发音序列；根据待检测发音序列构建待检测FST；根据待检测FST和关键词FST，确定待检测文本中的若干待纠错字词，并确定包含待纠错字词的待纠错语句；若汉字混淆集中存在待纠错字词，确定每个待纠错字词对应的替换字词，并将待纠错语句中的待纠错字词替换为替换字词，生成替换语句；当待纠错语句的第一逻辑得分小于替换语句的第二逻辑得分，将待检测文本中的待纠错语句替换为替换语句，完成纠错。本申请的技术方案可实现根据发音确定可能存在错误的字词，实现对语音识别文本的有效纠错。

9. 发明申请

WO2022266642A1 MULTILINGUAL GRAMMATICAL ERROR CORRECTION 审中-公开
公开(公告)号：WO2022266642A1
公开(公告)日：2022-12-22
申请号：PCT/US2022/072965
申请日：2022-06-15
申请人： GOOGLE LLC
发明人： KRAUSE, Sebastian , ROTHE, Sascha , MALLINSON, Jonathan , MALMI, Eric , SEVERYN, Aliaksei
IPC分类号： G06F40/232 , G06F40/253
摘要： A method (400) of training a text-generating model (122) for grammatical error correction (GEC) includes obtaining a multilingual set (210) of text samples (212). Each text sample includes a monolingual textual representation of a respective sentence. The operations also include, for each text sample of the multilingual set of text samples, generating a corrupted synthetic version (222) of the respective text sample, the corrupted synthetic version of the respective text sample including a grammatical change to the monolingual textual representation of the respective sentence associated with the respective text sample. The operations further include training the text-generating model using a training set (230) of sample pairs (232). Each sample pair includes one of the respective text samples of the multilingual set of text samples and the corresponding corrupted synthetic version of the one of the respective text samples of the multilingual set of text samples.

10. 发明申请

WO2021260554A1 DOMAIN-SPECIFIC GRAMMAR CORRECTION SYSTEM, SERVER AND METHOD FOR ACADEMIC TEXT 审中-公开
公开(公告)号：WO2021260554A1
公开(公告)日：2021-12-30
申请号：PCT/IB2021/055510
申请日：2021-06-22
申请人： CRIMSON AI LLP
发明人： NAGVENKAR, Apurva , RANE, Sachin , MITTAL, Sharad , D'CUNHA, Trinka
IPC分类号： G06F40/232 , G06F40/253 , G06F40/289 , G06F40/44 , G06F40/56 , G06N3/04
摘要： A method of identifying text (e.g., a sentence or sentence portion) in a word processing text editor; automatically identifying a domain-specific deep-learning neural network that corresponds to an identified context, from among one or more domain-specific deep-learning neural networks; automatically identifying at least one suggested replacement word using the identified domain specific deep-learning neural network that corresponds to the identified context; and automatically controlling a display to display a user interface that includes functionality that presents prompt information that includes the at least one suggested replacement word. Changes for errors that are common in academic papers written by non-native speakers may be suggested.

你已经成功收藏专利！

检索式保存成功!

IPRDB

热门服务

关于我们

友情链接

联系方式