专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

51. 发明申请

WO2003034281A1 METHOD AND APPARATUS TO PROVIDE A HIERARCHICAL INDEX FOR A LANGUAGE MODEL DATA STRUCTURE 审中-公开
标题翻译：为语言模型数据结构提供分层索引的方法和装置
公开(公告)号：WO2003034281A1
公开(公告)日：2003-04-24
申请号：PCT/RU2001/000431
申请日：2001-10-19
申请人： INTEL ZAO , KIBKALO, Alexandr Alexeevich , RYZHACHKIN, Ivan, P.
发明人： KIBKALO, Alexandr Alexeevich , RYZHACHKIN, Ivan, P.
IPC分类号： G06F17/30
CPC分类号： G06F17/30625 , G10L15/197
摘要： A method for storing bigram word indexes of a language model for a consecutive speech recognition system (200) is described. The bigram word indexes (321) are stored as a common two-byte base with a specific one-byte offset to significantly reduce storage requirements of the language model data file. In one embodiment the storage space required for storing the bigram word indexes (321) sequentially is compared to the storage space required to store the bigram word indexes as a common base with specific offset. The bigram word indexes (321) are then stored so as to minimize the size of the language model data file.
摘要翻译：描述了一种用于存储用于连续语音识别系统（200）的语言模型的双字词索引的方法。双字词索引（321）被存储为具有特定一字节偏移量的公共两字节基址，以显着减少语言模型数据文件的存储要求。在一个实施例中，将二进制字词索引（321）顺序存储所需的存储空间与存储双字词索引作为具有特定偏移量的公共基址所需的存储空间进行比较。然后存储双字词索引（321），以便最小化语言模型数据文件的大小。

52. 发明申请

WO2002071392A1 SPEECH RECOGNITION SYSTEM WITH MAXIMUM ENTROPY LANGUAGE MODELS 审中-公开
标题翻译：具有最大熵语言模型的语音识别系统
公开(公告)号：WO2002071392A1
公开(公告)日：2002-09-12
申请号：PCT/IB2002/000634
申请日：2002-03-05
申请人： KONINKLIJKE PHILIPS ELECTRONICS N.V. , PHILIPS CORPORATE INTELLECTUAL PROPERTY GMBH , PETERS, Jochen
发明人： PETERS, Jochen
IPC分类号： G10L15/18
CPC分类号： G10L15/197 , G10L15/183
摘要： The invention relates to a method of setting a free parameter of an attribute in a maximum-entropy speech model, which free parameter could not be set previously with the help of a training algorithm. It is an object of the invention to provide a speech recognition system 100, a training device 10 and a method of setting such a parameter that has a number of possible interpretations. This object is achieved in accordance with the invention in that is calculated as follows: formula (I) with formula (II) and denominator( = formula (III).
摘要翻译：本发明涉及一种在最大熵语音模型中设置属性的自由参数的方法，该自由参数在训练算法的帮助下不能先前设定。本发明的一个目的是提供一种语音识别系统100，训练装置10以及一种设置具有多个可能解释的参数的方法。该目的根据本发明实现，其计算公式如下：式（I）与式（II）和分母（=式（III））。

53. 发明申请

WO01037128A2 A SYSTEM AND ITERATIVE METHOD FOR LEXICON, SEGMENTATION AND LANGUAGE MODEL JOINT OPTIMIZATION 审中-公开
标题翻译：用于LEXICON，SEGMENTATION和语言模型联合优化的系统和迭代方法
公开(公告)号：WO01037128A2
公开(公告)日：2001-05-25
申请号：PCT/US2000/041870
申请日：2000-11-03
IPC分类号： G06F17/28 , G06F17/27 , G10L15/06 , G10L15/18 , G06F17/20
CPC分类号： G06F17/274 , G10L15/197
摘要： A method for optimizing a language model is presented comprising developing an initial language model from a lexicon and segmentation derived from a received corpus using a maximum match technique, and iteratively refining the initial language model by dynamically updating the lexicon and re-segmenting the corpus according to statistical principles until a threshold of predictive capability is achieved.
摘要翻译：提出了一种用于优化语言模型的方法，其包括使用最大匹配技术从词典和从接收到的语料库导出的分割开发初始语言模型，并且通过动态地更新词典并重新分割语料库来迭代地改进初始语言模型统计原理，直到达到预测能力的阈值。

54. 发明申请

WO00073936A1 PHRASE-BASED DIALOGUE MODELING WITH PARTICULAR APPLICATION TO CREATING RECOGNITION GRAMMARS FOR VOICE-CONTROLLED USER INTERFACES 审中-公开
标题翻译：基于PHRASE的对话建模，特别应用于语音控制用户界面创建识别格式
公开(公告)号：WO00073936A1
公开(公告)日：2000-12-07
申请号：PCT/US2000/014961
申请日：2000-05-27
IPC分类号： G06F17/28 , G06F17/27 , G10L13/00 , G10L15/18 , G10L15/22 , G10L19/00 , G06F17/21 , G10L21/00
CPC分类号： G06F17/271 , G06F17/2715 , G06F17/2765 , G06F17/2775 , G06F17/279 , G06F17/2795 , G10L15/005 , G10L15/183 , G10L15/193 , G10L15/197 , G10L15/22
摘要： The invention enables creation of grammar networks that can regulate, control, and define the content and scope of human-machine interaction in natural language voice user interfaces (NLVUI). More specifically, the invention concerns a phrase-based modeling of generic structures of verbal interaction and use of these models for the purpose of automating part of the design of such grammar networks.
摘要翻译：本发明能够创建可以调节，控制和定义自然语言语音用户界面（NLVUI）中人机交互的内容和范围的语法网络。更具体地，本发明涉及用于自动化这种语法网络的设计的一部分的目的语言交互的通用结构和这些模型的使用的基于短语的建模。

55. 发明申请

WO00023983A1 METHOD OF DETERMINING PARAMETERS OF A STATISTICAL LANGUAGE MODEL 审中-公开
标题翻译：确定统计语言模型参数的方法
公开(公告)号：WO00023983A1
公开(公告)日：2000-04-27
申请号：PCT/EP1999/007810
申请日：1999-10-11
IPC分类号： G06F17/28 , G10L15/183 , G10L15/197 , G10L15/18
CPC分类号： G10L15/197 , G10L15/183
摘要： The invention relates to a method of determining parameters of a statistical language model for automatic speech recognition systems by means of a training corpus. To improve the perplexity and the error rate in the speech recognition, it is proposed that at least a proportion of the elements of a vocabulary used is combined so as to form context-independent vocabulary element categories; the frequencies of occurrence of vocabulary element sequences, and if applicable the frequencies of occurrence of derived sequences formed from said vocabulary element sequences through the replacement of at least one vocabulary element by the associated vocabulary element class, are evaluated in the language modeling process; and the parameters of the language model are derived from the frequencies of occurrence thus determined.
摘要翻译：本发明涉及一种通过训练语料库确定自动语音识别系统的统计语言模型的参数的方法。为了提高语音识别中的困惑性和错误率，提出将所使用词汇的元素的至少一部分组合起来，以形成与上下文无关的词汇单元类别; 在语言建模过程中评估词汇元素序列的出现频率，以及如果适用的话，通过相关词汇元素类替换至少一个词汇单元从所述词汇单元序列形成的派生序列的出现频率; 并且语言模型的参数从如此确定的发生频率导出。

56. 发明申请

WO99022363A1 SELECTION OF SUPERWORDS BASED ON CRITERIA RELEVANT TO BOTH SPEECH RECOGNITION AND UNDERSTANDING 审中-公开
标题翻译：基于与两个语音识别相关的标准选择超级词汇和理解
公开(公告)号：WO99022363A1
公开(公告)日：1999-05-06
申请号：PCT/US1998/022399
申请日：1998-10-23
IPC分类号： G10L15/18 , G10L5/06
CPC分类号： G10L15/18 , G10L15/197
摘要： This invention is directed to the selection of superwords based on a criterion relevant to speech recognition and understanding. Superwords are used to refer to those word combinations which are so often spoken that they are recognized as units or should have models to reflect them in the language model. The selected superwords are placed in a lexicon along with selected meaningful phrases. The lexicon is then used by a speech recognizer to improve recognition of input speech utterances for the proper routing of a user's task objectives.
摘要翻译：本发明旨在基于与语音识别和理解相关的标准来选择词条。超级词用于指通常被说出来的单词组合，它们被认为是单位，或者应该具有在语言模型中反映它们的模型。所选的超级词汇与选定的有意义的短语一起放置在词典中。然后语音识别器使用词典来改善输入语音话语的识别，以便正确地路由用户的任务目标。

57. 发明申请

WO99014740A1 METHOD FOR DETERMINING THE PROBABILITY OF THE OCCURRENCE OF A SEQUENCE OF AT LEAST TWO WORDS IN A SPEECH RECOGNITION PROCESS 审中-公开
标题翻译：方法确定至少两个词序列的发生的概率语音识别
公开(公告)号：WO99014740A1
公开(公告)日：1999-03-25
申请号：PCT/DE1998/002632
申请日：1998-09-07
IPC分类号： G10L15/183 , G10L15/197 , G10L5/06
CPC分类号： G10L15/197 , G10L15/183
摘要： Statistical speech models are used to improve the recognition rate of automatic speech recognition systems. The inventive method makes it possible to construct speech models with small volumes of text or without a text. Linguistic classes are formed and the speech model, i.e. an a-priori probability of word sequences, is determined based on the grammatical structure of the text and depending on the vocabulary of the text. Set probabilities for words and for linguistic classes are determined for this purpose. In addition to creating a new speech model for a predetermined area of application, the inventive method also provides a means of adapting a pre-existing speech model to a new area of application.
摘要翻译：为了提高自动语音识别系统的检出率，使用统计语言模型。利用上述方法，可以对少量的文字或没有文字建立语言模型。有形成语言类和语言模型，被确定基于所述文本的语法结构的单词序列的一个先验概率取决于文本的词汇。这些单词和语言类的条件概率被确定。除了给定应用的语言模型的再创造，该方法还包括用于适配现有的语言模型到新的应用的可能性。

58. 发明申请

WO1993013519A1 COMPOSITE EXPERT 审中-公开
标题翻译：复合专家
公开(公告)号：WO1993013519A1
公开(公告)日：1993-07-08
申请号：PCT/US1992011375
申请日：1992-12-18
申请人： KURZWEIL APPLIED INTELLIGENCE, INC.
发明人： KURZWEIL APPLIED INTELLIGENCE, INC. , GANONG, William, F., III
IPC分类号： G10L09/00
CPC分类号： G10L15/18 , G10L15/197
摘要： In a continuous speech recognizer which includes at least one acoustic expert and one linguistic expert which generate respective scores, a method is disclosed for adjusting the relative weighting to be applied to those scores employing training data utilizing the words to be recognized in multiple word phrases. Multiple word test phrases are applied to the acoustic expert to determine, for each phrase, plural multi-word hypotheses (15) each having corresponding cumulative scores. The linguistic expert generates corresponding cumulative linguistic scores (17). An objective function is calculated (25) for each test phrase having a value which is variable as a function of the difference between the combined score of any correct hypothesis and that of the most easily confused incorrect hypothesis. The objective function values are cumulated (27) and a gradient descent procedure is used (31) to adjust the relative weighting of the acoustic and linguistic scores in obtaining a combined score.

59. 发明申请

WO2017215122A1 多语音识别模型切换方法、装置和存储介质审中-公开
公开(公告)号：WO2017215122A1
公开(公告)日：2017-12-21
申请号：PCT/CN2016/097417
申请日：2016-08-30
申请人：百度在线网络技术(北京)有限公司
发明人：蒋兵 , 李先刚 , 丁科
IPC分类号： G10L15/00
CPC分类号： G10L15/197 , G06F17/27 , G10L15/005 , G10L15/02 , G10L15/22 , G10L25/78
摘要：一种多语音识别模型切换方法、装置和存储介质，方法包括：获取用户输入语音中的至少一条语音信息(101)；对语音信息进行识别和语言类别的匹配，以根据匹配度确定对应的目标语言类别(102)；将当前使用的语音识别模型切换至与目标语言类别对应的语音识别模型(103)。通过对获取的语音信息进行识别和语言类别的匹配，以根据匹配度确定对应的目标语言类别；将当前使用的语音识别模型切换至与目标语言类别对应的语音识别模型，能够实现不同语言间的语音识别模型的自动切换，不仅提高语音识别模型切换效率，而且使语音识别更加智能化。

60. 发明申请

WO2017205639A1 METHOD AND SYSTEM FOR HYBRID DECODING FOR ENHANCED END-USER PRIVACY AND LOW LATENCY 审中-公开
标题翻译：对增强的最终用户隐私和低延迟进行混合解码的方法和系统
公开(公告)号：WO2017205639A1
公开(公告)日：2017-11-30
申请号：PCT/US2017/034504
申请日：2017-05-25
申请人： NUANCE COMMUNICATIONS, INC.
发明人： QUILLEN, Carl, Benjamin , PARIHAR, Naveen
IPC分类号： G10L15/32 , G10L15/30 , G10L15/197
CPC分类号： G10L15/30 , G10L15/14 , G10L15/18 , G10L15/197 , G10L15/32 , H04L63/04 , H04L63/0428
摘要： Methods described herein provide functionality for automatic speech recognition (ASR). One such embodiment performs speech recognition using received speech recognition result candidates, where the received candidates were generated by performing Statistical Language Model (SLM) based speech recognition on one or more frames of audio data. In turn, such an embodiment transmits results of the speech recognition, performed using the received speech recognition result candidates, to a user device via a communications network. Results of the speech recognition are available with lower latency than pure cloud based ASR solution.
摘要翻译：这里描述的方法提供了用于自动语音识别（ASR）的功能。一个这样的实施例使用接收到的语音识别结果候选来执行语音识别，其中所接收的候选是通过对一个或多个音频数据帧执行基于统计语言模型（SLM）的语音识别而生成的。进而，这样的实施例通过通信网络将使用接收到的语音识别结果候选执行的语音识别的结果发送到用户装置。语音识别的结果比基于纯云的ASR解决方案具有更低的延迟。

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