-
公开(公告)号：DE60312963D1
公开(公告)日：2007-05-16
申请号：DE60312963
申请日：2003-01-30
Applicant: PHILIPS INTELLECTUAL PROPERTY
Inventor： THELEN ERIC , KLAKOW DIETRICH , SCHOLL HOLGER , WAIBEL ULRICH , REISINGER JOSEF
Abstract: The invention relates to a method and a device for the transcription of spoken and written utterances. To this end, the utterances undergo speech or text recognition, and the recognition result (ME) is combined with a manually created transcription (MT) of the utterances in order to obtain the transcription. The additional information rendered usable by the combination as a result of the recognition result (ME) enables the transcriber to work relatively roughly and therefore quickly on the manual transcription. When using a keyboard ( 25 ), he can, for example, restrict himself to hitting the keys of only one row and/or can omit some keystrokes completely. In addition, the manual transcribing can also be accelerated by the suggestion of continuations ( 31 ) to the text input so far ( 30 ), which continuations are anticipated by virtue of the recognition result (ME).
-
公开(公告)号：DE10204924A1
公开(公告)日：2003-08-21
申请号：DE10204924
申请日：2002-02-07
Applicant: PHILIPS INTELLECTUAL PROPERTY
Inventor： THELEN ERIC , KLAKOW DIETRICH , SCHOLL HOLGER , WAIBEL ULI , REISINGER JOSEF
Abstract: The invention relates to a method and a device for the transcription of spoken and written utterances. To this end, the utterances undergo speech or text recognition, and the recognition result (ME) is combined with a manually created transcription (MT) of the utterances in order to obtain the transcription. The additional information rendered usable by the combination as a result of the recognition result (ME) enables the transcriber to work relatively roughly and therefore quickly on the manual transcription. When using a keyboard ( 25 ), he can, for example, restrict himself to hitting the keys of only one row and/or can omit some keystrokes completely. In addition, the manual transcribing can also be accelerated by the suggestion of continuations ( 31 ) to the text input so far ( 30 ), which continuations are anticipated by virtue of the recognition result (ME).
-
公开(公告)号：DE59912920D1
公开(公告)日：2006-01-19
申请号：DE59912920
申请日：1999-09-08
Applicant: PHILIPS INTELLECTUAL PROPERTY
Inventor： KLAKOW DIETRICH
IPC: G10L15/197
Abstract: The method involves raising different M-gram probabilities of an element to the power of M-gram-specifically optimised parameter values and multiplying the results. The estimate of probability does not include the case where a probability with M greater than 1 for a speech vocabulary element estimated with a training vocabulary body is multiplied by a quotient raised to the power of an optimised parameter value. The optimised parameter value is determined using a GIS algorithm with the dividend as a unigram probability estimated using a second training vocabulary body and a unigram probability estimated using the first training vocabulary body as divisor. An Independent claim is also included for a speech recognition system.
-
公开(公告)号：DE60312963T2
公开(公告)日：2007-12-13
申请号：DE60312963
申请日：2003-01-30
Applicant: PHILIPS INTELLECTUAL PROPERTY
Inventor： THELEN ERIC , KLAKOW DIETRICH , SCHOLL HOLGER , WAIBEL ULRICH , REISINGER JOSEF
Abstract: The invention relates to a method and a device for the transcription of spoken and written utterances. To this end, the utterances undergo speech or text recognition, and the recognition result (ME) is combined with a manually created transcription (MT) of the utterances in order to obtain the transcription. The additional information rendered usable by the combination as a result of the recognition result (ME) enables the transcriber to work relatively roughly and therefore quickly on the manual transcription. When using a keyboard ( 25 ), he can, for example, restrict himself to hitting the keys of only one row and/or can omit some keystrokes completely. In addition, the manual transcribing can also be accelerated by the suggestion of continuations ( 31 ) to the text input so far ( 30 ), which continuations are anticipated by virtue of the recognition result (ME).
-
公开(公告)号：DE69919842T2
公开(公告)日：2005-09-01
申请号：DE69919842
申请日：1999-12-16
Inventor： STEINBISS VOLKER , KLAKOW DIETRICH
IPC: G10L15/06 , G10L15/14 , G10L15/18 , G10L15/197 , G10L15/28
Abstract: A small vocabulary pattern recognition system is used for recognizing a sequence of words, such as a sequence of digits (e.g. telephone number) or a sequence of commands. A representation of reference words is stored in a vocabulary 132, 134. Input means 110 are used for receiving a time-sequential input pattern representative of a spoken or written word sequence. A pattern recognizer 120 comprises a word-level matching unit 130 for generating a plurality of possible sequences of words by statistically comparing the input pattern to the representations of the reference words of the vocabulary 132, 134. A cache 150 is used for storing a plurality of most recently recognized words. A sequence-level matching unit 140 selects a word sequence from the plurality of sequences of words in dependence on a statistical language model which provides a probability of a sequence of M words, M>=2. The probability depends on a frequency of occurrence of the sequence in the cache. In this way for many small vocabulary systems where no reliable data is available on frequency of use of word sequences, the cache is used to provide data representative of the actual use.
-
Patent Agency Ranking
公开(公告)号：DE69919842D1
公开(公告)日：2004-10-07
申请号：DE69919842
申请日：1999-12-16
Inventor： STEINBISS VOLKER , KLAKOW DIETRICH
IPC: G10L15/06 , G10L15/14 , G10L15/18 , G10L15/197 , G10L15/28
Abstract: A small vocabulary pattern recognition system is used for recognizing a sequence of words, such as a sequence of digits (e.g. telephone number) or a sequence of commands. A representation of reference words is stored in a vocabulary 132, 134. Input means 110 are used for receiving a time-sequential input pattern representative of a spoken or written word sequence. A pattern recognizer 120 comprises a word-level matching unit 130 for generating a plurality of possible sequences of words by statistically comparing the input pattern to the representations of the reference words of the vocabulary 132, 134. A cache 150 is used for storing a plurality of most recently recognized words. A sequence-level matching unit 140 selects a word sequence from the plurality of sequences of words in dependence on a statistical language model which provides a probability of a sequence of M words, M>=2. The probability depends on a frequency of occurrence of the sequence in the cache. In this way for many small vocabulary systems where no reliable data is available on frequency of use of word sequences, the cache is used to provide data representative of the actual use.
-
公开(公告)号：AU2003205955A1
公开(公告)日：2003-09-02
申请号：AU2003205955
申请日：2003-01-30
Applicant: PHILIPS INTELLECTUAL PROPERTY
Inventor： THELEN ERIC , KLAKOW DIETRICH , SCHOLL HOLGER , WAIBEL ULRICH , REISINGER JOSEF
Abstract: The invention relates to a method and a device for the transcription of spoken and written utterances. To this end, the utterances undergo speech or text recognition, and the recognition result (ME) is combined with a manually created transcription (MT) of the utterances in order to obtain the transcription. The additional information rendered usable by the combination as a result of the recognition result (ME) enables the transcriber to work relatively roughly and therefore quickly on the manual transcription. When using a keyboard ( 25 ), he can, for example, restrict himself to hitting the keys of only one row and/or can omit some keystrokes completely. In addition, the manual transcribing can also be accelerated by the suggestion of continuations ( 31 ) to the text input so far ( 30 ), which continuations are anticipated by virtue of the recognition result (ME).
-
8.发明申请METHOD FOR STORING BROADCAST CONTENTS, AND A BROADCAST CONTENT STORAGE SYSTEM 审中-公开存储广播内容的方法和广播内容存储系统
公开(公告)号：WO2005101708A2
公开(公告)日：2005-10-27
申请号：PCT/IB2005051125
申请日：2005-04-06
Applicant: PHILIPS INTELLECTUAL PROPERTY , KONINKL PHILIPS ELECTRONICS NV , THELEN ERIC , KLAKOW DIETRICH
Inventor： THELEN ERIC , KLAKOW DIETRICH
CPC classification number: H04H60/73 , G11B27/034 , G11B27/105 , G11B2220/2516 , H04H60/27 , H04H60/47 , H04N5/781 , H04N5/782
Abstract: The invention describes a method for storing broadcast contents and a broadcast content storage system. A plurality of content categories (KAT1, KAT2) is pre-defined, each of which is defined or described by at least one content descriptor (OB 1, OB2). Broadcast contents transmitted over at least one broadcast transmission channel are received, preferably continually, or over pre-defined lengths of time. Received broadcast contents, which are described by a content descriptor (OB1, OB2), are automatically assigned to the content category (KAT1, KAT2) defined or described by the corresponding content descriptor (OB1, OB2). The broadcast contents assigned to a content category (KAT1, KAT2) and the assignments of the broadcast contents to the corresponding content categories (KAT1, KAT2) are automatically stored.
Abstract translation: 本发明描述了一种用于存储广播内容的方法和广播内容存储系统。 多个内容类别（KAT1，KAT2）被预定义，每个内容类别由至少一个内容描述符（OB 1，OB 2）定义或描述。 通过至少一个广播传输信道发送的广播内容优选地连续地或超过预定义的时间长度被接收。 由内容描述符（OB1，OB2）描述的接收的广播内容被自动分配给由对应的内容描述符（OB1，OB2）定义或描述的内容类别（KAT1，KAT2）。 自动存储分配给内容类别（KAT1，KAT2）的广播内容和广播内容对相应内容类别（KAT1，KAT2）的分配。
-
公开(公告)号：WO2005050621A3
公开(公告)日：2005-10-27
申请号：PCT/IB2004052403
申请日：2004-11-12
Applicant: PHILIPS INTELLECTUAL PROPERTY , KONINKL PHILIPS ELECTRONICS NV , PETERS JOCHEN , MATUSOV EVGENY , MEYER CARSTEN , KLAKOW DIETRICH
Inventor： PETERS JOCHEN , MATUSOV EVGENY , MEYER CARSTEN , KLAKOW DIETRICH
CPC classification number: G10L15/183 , G06F17/211 , G06F17/2715 , G10L15/32
Abstract: The present invention relates to a method, a computer system and a computer program product for speech recognition and/or text formatting by making use of topic specific statistical models. A text document which may be obtained from a first speech recognition pass is subject to segmentation and to an assignment of topic specific models for each obtained section. Each model of the set of models provides statistic information about language model probabilities, about text processing or formatting rules, as e.g. the interpretation of commands for punctuation, formatting, text highlighting or of ambiguous text portions requiring specific formatting, as well as a specific vocabulary being characteristic for each section of the recognized text. Furthermore, other properties of a speech recognition and/or formatting system (such as e.g. settings for the speaking rate) may be encoded in the statistical models. The models themselves are generated on the basis of annotated training data and/or by manual coding. Based on the assignment of models to sections of text an improved speech recognition and/or text formatting procedure is performed.
Abstract translation: 本发明涉及通过利用话题特定的统计模型来进行语音识别和/或文本格式化的方法，计算机系统和计算机程序产品。 可以从第一语音识别过程中获得的文本文档受到分割以及为每个获得的章节分配主题特定模型。 该组模型的每个模型提供关于语言模型概率的统计信息，例如关于文本处理或格式化规则，例如， 用于标点符号，格式化，文本突出显示或需要特定格式的模糊文本部分的命令的解释，以及针对识别文本的每个部分的特定词汇表。 此外，可以在统计模型中编码语音识别和/或格式化系统的其他属性（例如，对于讲话速率的设置）。 模型本身是基于注释的训练数据和/或手动编码生成的。 基于将模型分配给文本段，执行改进的语音识别和/或文本格式化过程。
-
公开(公告)号：WO2005050472A3
公开(公告)日：2006-07-20
申请号：PCT/IB2004052404
申请日：2004-11-12
Applicant: PHILIPS INTELLECTUAL PROPERTY , KONINKL PHILIPS ELECTRONICS NV , PETERS JOCHEN , MEYER CARSTEN , KLAKOW DIETRICH , MATUSOV EVGENY
Inventor： PETERS JOCHEN , MEYER CARSTEN , KLAKOW DIETRICH , MATUSOV EVGENY
CPC classification number: G06F17/27 , G06F17/2765
Abstract: The invention relates to a method, a computer program product and a computer system for structuring an unstructured text by making use of statistical models trained on annotated training data. Each section of text in which the text is segmented is further assigned to a topic which is associated to a set of labels. The statistical models for the segmentation of the text and for the assignment of a topic and its associated labels to a section of text explicitly accounts for: correlations between a section of text and a topic, a topic transition between sections, a topic position within the document and a (topic-dependent) section length. Hence structural information of the training data is exploited in order to perform segmentation and annotation of unknown text.
Abstract translation: 本发明涉及一种通过利用在注释训练数据上训练的统计模型来构造非结构化文本的方法，计算机程序产品和计算机系统。 将文本分割的文本的每个部分进一步分配给与一组标签相关联的主题。 用于文本分段和用于将主题及其关联标签分配给文本部分的统计模型明确地表示：文本部分与主题之间的相关性，部分之间的主题转换，内容中的主题位置 文件和（主题相关）部分长度。 因此，利用训练数据的结构信息来执行未知文本的分割和注释。
-
-
-
-
-
-
-
-
-
Search Results
11
records
(took 0.023 seconds)
