Abstract: An adaptive dialogue system and also a computer-implemented method for semantic training of a dialogue system are disclosed. In this connection, semantic annotations are generated automatically on the basis of received speech inputs, the semantic annotations being intended for controlling instruments or for communication with a user. For this purpose, at least one speech input is received in the course of an interaction with a user. A sense content of the speech input is registered and appraised, by the speech input being classified on the basis of a trainable semantic model, in order to make a semantic annotation available for the speech input. Further user information connected with the speech input is taken into account if the registered sense content is appraised erroneously, incompletely and/or as untrustworthy. The sense content of the speech input is learned automatically on the basis of the additional user information.

Abstract: An adaptive dialogue system and also a computer-implemented method for semantic training of a dialogue system are disclosed. In this connection, semantic annotations are generated automatically on the basis of received speech inputs, the semantic annotations being intended for controlling instruments or for communication with a user. For this purpose, at least one speech input is received in the course of an interaction with a user. A sense content of the speech input is registered and appraised, by the speech input being classified on the basis of a trainable semantic model, in order to make a semantic annotation available for the speech input. Further user information connected with the speech input is taken into account if the registered sense content is appraised erroneously, incompletely and/or as untrustworthy. The sense content of the speech input is learned automatically on the basis of the additional user information.

Abstract: The present invention discloses an apparatus for automatic extraction of important events in audio signals comprising: signal input means for supplying audio signals; audio signal fragmenting means for partitioning audio signals supplied by the signal input means into audio fragments of a predetermined length and for allocating a sequence of one or more audio fragments to a respective audio window; feature extracting means for analyzing acoustic characteristics of the audio signals comprised in the audio fragments and for analyzing acoustic characteristics of the audio signals comprised in the audio windows; and important event extraction means for extracting important events in audio signals supplied by the audio signal fragmenting means based on predetermined important event classifying rules depending on acoustic characteristics of the audio signals comprised in the audio fragments and on acoustic characteristics of the audio signals comprised in the audio windows, wherein each important event extracted

Abstract: A user profile and/or the suggestions computed based thereon are obtained taking a special set of user features into account. The user features are defined to represent a typical general behaviour of an individual user in respect to the application where the user profile is used. In other words, for each application where a user profile is used a special set of user features are defined which are able to represent a typical general behaviour of an individual user. Based on these user features the weights in the list of word-weight pairs or weighted keywords which represents the user profile are computed or influenced during the creation of the user profile, and/or a mufti-user profile is split during the creation of an individual user profile from a mufti-user profile, and/or during specification of a suggestion a user history which is used to create the user profile, and/or the user profile, and/or the suggestion results are filtered.

Abstract: An apparatus for automatic dissection of segmented audio signals, wherein at least one information signal for identifying programs included in said audio signals and for identifying contents included in said programs. Content detection device detects programs and contents belonging to the respective programs in the information signal. Program weighting device weights each program includes in the information signal based on the contents of the respective program detected by the content detection device. Program ranking device indentifies programmers of the same category and ranking said programs based on a weighting result for each program provided by the program weighting device.

Abstract: An audio data segmentation apparatus for segmenting of audio data including for supplying audio data, dividing the audio data supplied into audio clips of a predetermined length, discriminating the audio clips into predetermined audio classes, the audio classes identifying a kind of audio data included in the respective audio clip and segmenting for segmenting the audio data into audio meta patterns based on a sequence of audio classes of consecutive audio clips, each meta pattern being allocated to a predetermined type of contents of the audio data. It is difficult to achieve good results with known methods for segmentation of audio data into meta patterns since the rules for the allocation of the meta patterns are dissatisfying.

Abstract: A user profile and/or the suggestions computed based thereon are obtained taking a special set of user features into account. The user features are defined to represent a typical general behaviour of an individual user in respect to the application where the user profile is used. In other words, for each application where a user profile is used a special set of user features are defined which are able to represent a typical general behaviour of an individual user. Based on these user features the weights in the list of word-weight pairs or weighted keywords which represents the user profile are computed or influenced during the creation of the user profile, and/or a mufti-user profile is split during the creation of an individual user profile from a mufti-user profile, and/or during specification of a suggestion a user history which is used to create the user profile, and/or the user profile, and/or the suggestion results are filtered.

Abstract: A user profile and/or the suggestions computed based thereon are obtained taking a special set of user features into account. The user features are defined to represent a typical general behaviour of an individual user in respect to the application where the user profile is used. In other words, for each application where a user profile is used a special set of user features are defined which are able to represent a typical general behaviour of an individual user. Based on these user features the weights in the list of word-weight pairs or weighted keywords which represents the user profile are computed or influenced during the creation of the user profile, and/or a multi-user profile is split during the creation of an individual user profile from a multi-user profile, and/or during specification of a suggestion a user history which is used to create the user profile, and/or the user profile, and/or the suggestion results are filtered.

Abstract: To increase the recognition rate and quality in a process of recognizing speech an approximative set of pronunciation rules (APR) for a current pronunciation (CP) of a current speaker is determined in a given pronunciation space (PS) and then applied to a current pronunciation lexicon (CL) so as to perform a speaker specific adaptation of said current lexicon (CL).

Abstract: A user profile and/or the suggestions computed based thereon are obtained taking a special set of user features into account. The user features are defined to represent a typical general behaviour of an individual user in respect to the application where the user profile is used. In other words, for each application where a user profile is used a special set of user features are defined which are able to represent a typical general behaviour of an individual user. Based on these user features the weights in the list of word-weight pairs or weighted keywords which represents the user profile are computed or influenced during the creation of the user profile, and/or a multi-user profile is split during the creation of an individual user profile from a multi-user profile, and/or during specification of a suggestion a user history which is used to create the user profile, and/or the user profile, and/or the suggestion results are filtered.

Abstract: A method for recognizing speech is proposed wherein the process of recognition is started using the starting acoustic model (SAM) and wherein the current acoustic model (CAM) is modified by removing or cancelling model function mixture components (MFMjk) which are negligible for the description of the speaking behavior and quality of the current speaker. Therefore, the size of the acoustic model (SAM, CAM) is reduced by adaptation to the current speaker enabling fast performance and increased recognition efficiency.

Abstract: Apparatus and method for automatic dissection of segmented audio signals According to the present invention, an apparatus for automatic dissection of segmented audio signals, wherein at least one information signal for identifying programmes included in said audio signals and for identifying contents included in said programmes is provided, comprises: content detection means for detecting programmes and contents belonging to the respective programmes in the information signal; programme weighting means for weighting each programme comprised in the information signal based on the contents of the respective programme detected by the content detection means; and programme ranking means for identifying programmes of the same category and ranking said programmes based on a weighting result for each programme provided by the programme weighting means.

Abstract: An apparatus for classifying audio signals comprises audio signal clipping means for partitioning audio signals into audio clips, and class discrimination means for discriminating the audio clips provided by the audio signal clipping means into predetermined audio classes based on predetermined audio class classifying rules, by analysing acoustic characteristics of the audio signals comprised in the audio clips, wherein a predetermined audio class classifying rule is provided for each audio class, and each audio class represents a respective kind of audio signals comprised in the corresponding audio clip. The determination process to find acceptable audio class classifying rules for each audio class according to the prior art is depending on both the used raw audio signals and the personal experience of the person conducting the determination process. Thus, the determination process usually is very difficult, time consuming and subjective.

Abstract: A method and an apparatus for effecting the method are proposed that allow to define a subset of video signals from a source set of video signals on the basis of meta data available for the source set of video signals. The meta data assign a generic term to a sub-section of the audio channel of the source set of video signals, a class description to one or more sub-units of the sub-section for classifying the origin of the respective sub-unit, a category allocation to a segment, which is formed by a string of one or more classified sub-units of a sub-section, and a rating value to the segment for rating the reliability of the category allocation of the segment. The method includes steps for selecting segments of a sub-section with a rating value above a defined threshold value, assigning a priority value to each category, and specifying a first subset of video signals by defining an arrangement of selected segments by an order based on the respective priority and rating values related to each segment.

Abstract: An audio data segmentation apparatus for segmenting of audio data comprises audio data input means for supplying audio data, audio data clipping means for dividing the audio data supplied by the audio data input means into audio clips of a predetermined length, class discrimination means for discriminating the audio clips supplied by the audio data clipping means into predetermined audio classes, the audio classes identifying a kind of audio data included in the respective audio clip and segmenting means for segmenting the audio data into audio meta patterns based on a sequence of audio classes of consecutive audio clips, each meta pattern being allocated to a predetermined type of contents of the audio data. It is difficult to achieve good results with known methods for segmentation of audio data into meta patterns since the rules for the allocation of the meta patterns are dissatisfying.

Abstract: The present invention discloses an apparatus for automatic extraction of important events in audio signals comprising: signal input means for supplying audio signals; audio signal fragmenting means for partitioning audio signals supplied by the signal input means into audio fragments of a predetermined length and for allocating a sequence of one or more audio fragments to a respective audio window; feature extracting means for analysing acoustic characteristics of the audio signals comprised in the audio fragments and for analysing acoustic characteristics of the audio signals comprised in the audio windows; and important event extraction means for extracting important events in audio signals supplied by the audio signal fragmenting means based on predetermined important event classifying rules depending on acoustic characteristics of the audio signals comprised in the audio fragments and on acoustic characteristics of the audio signals comprised in the audio windows, wherein each important event extracted by

Abstract: To avoid an over-adaptation of a current acoustic model (CAM) to certain and frequently occuring words for speech phrases during on-line speaker adaptation of speech recognizers it is suggested to count adaptation numbers (aj) for each of said speech phrases (SPj) as numbers of times in that a distinct speech phrase (SPj) has been used as a basis for adapting said current acoustic model (CAM) and further to make the strength of adaptation of the current acoustic model (CAM) on the basis of said distinct speech phrase (SPj) dependent on its specific adaptation number (aj) so as to decrease the influence of frequent speech phrases (SPj) in the received speech flow on the adaptation process.

Abstract: A method for recognizing speech comprising the steps of receiving a speech input (SI) of a user, determining a set of ordered hypotheses (OH) for said received speech input (SI), wherein said set of ordered hypotheses (OH) contains tag information (TI) for each of said ordered hypotheses, which is descriptive for at least one type or variation of pronunciation, using a tag language model (LM2) operating on said tag information (TI), re-ordering said set of hypotheses using said tag language model (LM2), outputting a set of re-ordered hypotheses (ROH) and choosing the best hypothesis (BH).

Abstract: To prevent adaptation to misrecognized words in unsupervised or on-line automatic speech recognition systems confidence measures are used or the user reaction is interpreted to decide whether a recognized phoneme, several phonemes, a word, several words or a whole utterance should be used for adaptation of the speaker independent model set to a speaker adapted model set or not and, in case an adaptation is executed, how strong the adaptation with this recognized utterance or part of this recognized utterance should be performed. Furtheron, a verification of the speaker adaptation performance is proposed to secure that the recognition rate never decreases (significantly), but only increases or stays at the same level.

Abstract: According to the present invention network devices that can be controlled via a speech unit included in the network can send a device-document describing its functionality and its speech interface to said speech unit. The speech unit combines those documents to a general document that forms the basis to translate recognized user-commands into user-network-commands to control the connected network-devices. A device-document comprises at least the vocabulary and the commands associated therewith for the corresponding device. Furtheron, pronunciation, grammar for word sequences, rules for speech understanding and dialog can be contained in such documents as well as the same information for multiple languages or information for dynamic dialogs in speech understanding. It is possible that one device contains several documents and dynamically sends them to the speech unit in case they are needed.