Abstract: A system facilitates the browsing of information of interest. The system obtains a transcription of the information and provides the transcription to a user. The system also retrieves the information in its original format and presents the information to the user in the original format.

Abstract: A system improves recognition results. The system receives multimedia data and recognizes the multimedia data based on training data to generate documents. The system receives user augmentation relating to one of the documents or new documents from a user. The system supplements the training data with the user augmentation or new documents and retrains based on the supplemented training data.

Abstract: A system improves recognition results. The system receives multimedia data and recognizes the multimedia data based on training data to generate documents. The system receives user augmentation relating to one of the documents or new documents from a user. The system supplements the training data with the user augmentation or new documents and retrains based on the supplemented training data.

Abstract: A system improves recognition results. The system receives multimedia data and recognizes the multimedia data based on training data to generate documents. The system receives user augmentation relating to one of the documents or new documents from a user. The system supplements the training data with the user augmentation or new documents and retrains based on the supplemented training data.

Abstract: An audio classification system classifies sounds in an audio stream as belonging to one of a relatively small number of classes. The audio classification system includes a signal analysis component [301] and a decoder [302]. The decoder [302] includes a number of models [310-316] for performing the audio classifications. In one implementation, the possible classifications include: vowels, fricatives, narrowband, wideband, coughing, gender, and silence. The classified audio may be used to enhance speech recognition of the audio stream.

Abstract: A unified clustering tree (500) generates phoneme clusters based on an input sequence of phonemes. The number of possible clusters is significantly less than the number of possible combinations of input phonemes. Nodes (510, 511) in the unified clustering tree are arranged into levels such that the clustering tree generates clusters for multiple speech recognition models. Models that correspond to higher levels in the unified clustering tree are coarse models relative to more fine-grain models at lower levels of the clustering tree.

Abstract: A speech recognition system receives an audio signal and detects various features of the audio signal. For example, the system classifies the audio signal into speech and non-speech portions, genders of speakers corresponding to the speech portions, and channel bandwidths used by the speakers. The system detects speaker turns based on changes in the speakers and assigns labels to the speaker turns. The system verifies the genders of the speakers and the channel bandwidths used by the speakers and identifies one or more languages associated with the audio signal. The system recognizes the speech portions of the audio signal based on the various features of the audio signal.

Abstract: A system facilitates the searching and retrieval of multimedia data items. The system receives data items from different types of media sources and identifies regions in the data items. The regions include document regions, section regions, and passage regions. Each of the section regions corresponds to one of the document regions and each of the passage regions corresponds to one of the section regions and one of the document regions. The system stores document identifiers that relate to the document regions in separate document records in a document table, section identifiers that relate to the section regions in separate section records in a section table, and passage identifiers that relate to the passage regions in separate passage records in a passage table.

Abstract: An audio classification system classifies sounds in an audio stream as belonging to one of a relatively small number of classes. The audio classification system includes a signal analysis component [301] and a decoder [302]. The decoder [302] includes a number of models [310-316] for performing the audio classifications. In one implementation, the possible classifications include: vowels, fricatives, narrowband, wideband, coughing, gender, and silence. The classified audio may be used to enhance speech recognition of the audio stream.

Abstract: A speaker change detection system performs speaker change detection on an input audio stream. The speaker change detection system includes a segmentation component [401], a phone classification decode component [402], and a speaker change detection component [403]. The segmentation component [401] segments the audio stream into segments [501-504] of predetermined length intervals. The segments may overlap one another. The phone classification decode component decodes the intervals to produce a set of phone classes corresponding to each of the intervals. The speaker change detection component detects locations of speaker changes in the audio stream based on a similarity value calculated at phone class boundaries.

Abstract: A system provides document browsing by proper name. The system identifies a subset of documents from a set of documents (310). The documents in the subset of documents include proper names. The system receives a selection of at least one of the proper names from the subset of documents (330) and searches the subset of documents to identify one or more of the documents in the subset of documents that include at least one occurrence of the selected proper name(s) (360). The system then presents the one or more of the documents as a result of the search (370).

Abstract: A dictionary creation component [316] converts the normal orthographic written representation of a word into a sequence of symbols that relate to the pronunciation of the word. The symbols may be used to train conventional models for speech recognition. The symbols are not phonemes and do not need to be defined by a speech expert. The symbols are created automatically by the dictionary creation component based on the written representation of the word.

Abstract: A system (520) generates labels for clusters of documents. The system (520) identifies topics associated with the documents in the clusters and determines whether the topics are associated with approximately half or more of the documents in the clusters. The system (520) then generates labels for the clusters using the topics that are associated with approximately half or more of the documents in the clusters.

Abstract: A transcription tool [115] includes a graphical user interface [209] that displays the waveform of an input audio signal to a user. The user may define speaker turn segments using the displayed waveform. The graphical user interface further displays a transcription section [302] that includes a textual representation of speech that was transcribed by the user and a graphical representation of annotation information [314] relating to the transcribed text. The user may enter the annotation information on-the-fly while transcribing the text using predefined keyboard shortcut commands or other mechanisms. The graphical user interface may further display a structured representation section [303] that may present the transcribed text as a hierarchical tree structure.

Abstract: A system (100) provides speaker identification training. The system (100) generates speaker models and receives audio segments. The system (100) identifies speakers corresponding to the audio segments based on the speaker models. At least one of the audio segments has an unidentified or misidentified speaker (i.e., an audio segment whose speaker cannot be accurately identified). The system (100) presents, to a user, audio segments that include an audio segment whose speaker is unidentified or misidentified and receives, from the user, the name of the unidentified or misidentified speaker. The system (100) may use this information to subsequently identify the unidentified or misidentified speaker by name for future audio segments.

Abstract: A speech system includes a Language Component Manager (LCM) [116] that functions as middleware between language and technology components [117] and a high-level application [115]. In aggregate, the LCM, the language and technology component, and the high-level application form a speech system. The components of the speech system may be distributed. All data and messages in the speech system pass through the LCM. A configuration file defines relationships, such as message and data paths, for the speech system.

Abstract: A speech recognition system receives an audio signal and detects various features of the audio signal. For example, the system classifies the audio signal into speech and non-speech portions, genders of speakers corresponding to the speech portions, and channel bandwidths used by the speakers. The system detects speaker turns based on changes in the speakers and assigns labels to the speaker turns. The system verifies the genders of the speakers and the channel bandwidths used by the speakers and identifies one or more languages associated with the audio signal. The system recognizes the speech portions of the audio signal based on the various features of the audio signal.

Abstract: A system facilitates the browsing of information of interest. The system obtains a transcription of the information and provides the transcription to a user. The system also retrieves the information in its original format and presents the information to the user in the original format. The system visually synchronizes the presentation of the information in the original format with the transcription of the information.

Abstract: A system improves recognition results. The system receives multimedia data and recognizes the multimedia data based on training data to generate documents. The system receives user augmentation relating to one of the documents or new documents from a user. The system supplements the training data with the user augmentation or new documents and retrains based on the supplemented training data.

Abstract: A system notifies a user of the detection of data that is relevant to an event of interest. The system obtains a user profile that includes one or more example documents that define the event. The system receives data that corresponds to multimedia information and determines the relevance of the data to the event based on the one or more example documents. The system notifies the user when the data is determined to be relevant.