2724 results about "Natural language processing" patented technology

The present invention relates to a method and system for textual exploration and discovery. More specifically, the method and system provide a text-driven and grammar based tool for textual exploration and textual navigation. The facilities for textual exploration and textual navigation are based on a system of index entries that are connected to the underlying text segments from which the index entries are derived. Text units with particular grammatical, semantic, and/or pragmatic features constitute bundles of sentences or text zones.

A system for retrieval of text includes a processor which identifies grammar rules associated with text fragments of a text string that is retrieved from an associated storage medium, and retrieves text strings from the storage medium which satisfy the grammar rules. A display displays retrieved text strings. A user input device in communication with the processor enables a user to select text fragments of the displayed text strings for generating a query. The processor identifies grammar rules associated with the user-selected text fragments and retrieves text strings from the storage medium which satisfy the grammar rules.

A knowledge-based natural speech dialogue system includes: (i) a knowledge support system, (ii) a flexible dialogue control system, and (iii) a context information system. Flexibilities of the conversation structure, inherent in mixed-initiative mode for dealing with complex user request, are managed because the knowledge structures involved are represented by additional, powerful knowledge representation tools, and because the context information is retained by more specific data structures, which covers larger temporal scopes by the logic of the conversation, rather than by a fixed locality of the grammar flow. This system provides a simple yet reliable method to compensate for these factors to enable more powerful conversation engines with mixed-initiative capabilities.

Systems and methods enhance editing capabilities associated with a wide variety of different types of electronic documents. Such systems and methods may include a processor that maintains an electronic document having a first portion (e.g., an individual word, character, character string, or the like) provided by a recognizer (e.g., by handwriting or speech recognition software), and they may provide access to potential alternative characters, words, or character strings generated by the recognizer during various user editing functions and operations. Other data associated with the first portion of the document also may be stored and made available to the user during various functions and operations. This invention further relates to computer-readable media including instructions for performing various methods and/or operating various systems for editing electronic documents, including systems and methods like those described above.

An apparatus for providing dynamic vocabulary prediction for setting up a speech recognition network of resource constrained portable devices may include a recognition network element. The recognition network element may be configured to determine a confidence measure for each candidate recognized word for a current word to be recognized. The recognition network element may also be configured to select a subset of candidate recognized words as selected candidate words based on the confidence measure of each one of the candidate recognized words, and determine a recognition network for a next word to be recognized, the recognition network including likely follower words for each of the selected candidate words, e.g. using language model and highly frequently used words.

Algorithms for synthesizing speech used to identify media assets are provided. Speech may be selectively synthesized form text strings associated with media assets. A text string may be normalized and its native language determined for obtaining a target phoneme for providing human-sounding speech in a language (e.g., dialect or accent) that is familiar to a user. The algorithms may be implemented on a system including several dedicated render engines. The system may be part of a back end coupled to a front end including storage for media assets and associated synthesized speech, and a request processor for receiving and processing requests that result in providing the synthesized speech. The front end may communicate media assets and associated synthesized speech content over a network to host devices coupled to portable electronic devices on which the media assets and synthesized speech are played back.

Algorithms for synthesizing speech used to identify media assets are provided. Speech may be selectively synthesized form text strings associated with media assets. A text string may be normalized and its native language determined for obtaining a target phoneme for providing human-sounding speech in a language (e.g., dialect or accent) that is familiar to a user. The algorithms may be implemented on a system including several dedicated render engines. The system may be part of a back end coupled to a front end including storage for media assets and associated synthesized speech, and a request processor for receiving and processing requests that result in providing the synthesized speech. The front end may communicate media assets and associated synthesized speech content over a network to host devices coupled to portable electronic devices on which the media assets and synthesized speech are played back.

A system and method are provided for improved speech synthesis, wherein text data is pre-processed according to updated grammar rules or a selected group of grammar rules. In one embodiment, the TTS system comprises a first memory adapted to store a text information database, a second memory adapted to store grammar rules, and a receiver adapted to receive update data regarding the grammar rules. The system also includes a TTS engine adapted to retrieve at least one text entry from the text information database, pre-process the at least one text entry by applying the updated grammar rules to the at least one text entry, and generate speech based at least in part on the least one pre-processed text entry.

Method and apparatus for correction of spelling errors in text composition is provided. Correctly spelled words may be determined for an incorrectly spelled word in accordance with a common error approach. Corrections to the incorrectly word are applied using common typographical and spelling errors. Resulting words which are correctly spelled may be offered as replacements or automatically used to replace the incorrectly word. Corrections may be applied in response to a frequency of the error in text composition to generate correct words that are more likely to be the word intended to be composed. The specific order of application may be configurable to meet a user's preference or a learned behavior for a user.

A recognition tool according to various examples of the invention intelligently recognizes natural input before it is passed to a destination or target application. More particularly, the recognition tool according to various examples of the invention provides better formatting for text recognized from natural input, based upon the context in which the text is being inserted into a target application. The recognition tool also provides various tools for correcting inaccurately recognized text. The recognition tool may allow a user to select only a part of an inaccurate text, and then identify alternate text candidates based upon the selected portion of the inaccurate text. Further, when the user selects text containing multiple words for correction, the recognition tool provides cross combinations of alternate text candidates for the user's selection. Still further, if the user replaces inaccurate text by submitting a new natural input object, the recognition tool ensures that the text recognized from the new natural input object is different from the inaccurate text been replaced. The recognition tool additionally affects the recognition experience after recognized text has been provided to the target application. The recognition tool provides the target application with the original natural input object for the recognized text, along with the alternate text candidates for that original natural input object. Thus, the target application can use the alternate text candidates to correct inaccurately recognized text. Further, a user can insert the original natural input object for recognized text within the target application.

Machine translation decoding is accomplished by receiving as input a text segment in a source language to be translated into a target language, generating an initial translation as a current target language translation, applying one or more modification operators to the current target language translation to generate one or more modified target language translations, determining whether one or more of the modified target language translations represents an improved translation in comparison with the current target language translation, setting a modified target language translation as the current target language translation, and repeating these steps until occurrence of a termination condition. Automatically generating a tree (e.g., either a syntactic tree or a discourse tree) can be accomplished by receiving as input a tree corresponding to a source language text segment, and applying one or more decision rules to the received input to generate a tree corresponding to a target language text segment.

Described is a speech recognition dialog management system that allows more open-ended conversations between virtual agents and people than are possible using just agent-directed dialogs. The system uses both novel dialog context switching and learning algorithms based on spoken interactions with people. The context switching is performed through processing multiple dialog goals in a last-in-first-out (LIFO) pattern. The recognition accuracy for these new flexible conversations is improved through automated learning from processing errors and addition of new grammars.

Systems and methods to process and correct spelling errors for non-Roman based words such as in Chinese, Japanese, and Korean languages using a rule-based classifier and a hidden Markov model are disclosed. The method generally includes converting an input entry in a first language such as Chinese to at least one intermediate entry in an intermediate representation, such as pinyin, different from the first language, converting the intermediate entry to at least one possible alternative spelling or form of the input in the first language, and determining that the input entry is either a correct or questionable input entry when a match between the input entry and all possible alternative spellings to the input entry is or is not located, respectively. The questionable input entry may be classified using, for example, a transformation rule based classifier based on transformation rules generated by a transformation rules generator.

An automated method is described for developing an automated speech input semantic classification system such as a call routing system. A set of semantic classifications is defined for classification of input speech utterances, where each semantic classification represents a specific semantic classification of the speech input. The semantic classification system is trained from training data having little or no in-domain manually transcribed training data, and then operated to assign input speech utterances to the defined semantic classifications. Adaptation training data based on input speech utterances is collected with manually assigned semantic labels. When the adaptation training data satisfies a pre-determined adaptation criteria, the semantic classification system is automatically retrained based on the adaptation training data.

A system that detects incorrectly translated text in a document. During operation, the system determines a target language for the document. The system then identifies one or more terms in the document that are invalid terms for the language. Next, the system analyzes correlations between these invalid terms, and, based on these correlations, determines whether one or more of the invalid terms comprise incorrectly translated text.

A method for transliterating languages in a computer system is disclosed. The method includes using a phonetic mapping engine to map languages to phonetic strings and vice versa. A user unfamiliar with a keyboard layout for a particular language may type phonetic strings using a known keyboard layout of a language known to the user. The typed input is captured by a keyboard hook, and based on a predefined phonetic mapping scheme, can be converted to the desired language.

An apparatus, method, computer system, and computer program product to optimize space available for displaying text descriptions in a limited area, such as a sidebar. Space is optimized by displaying only the maximum portion of the text describing each selection that will fit within the limited area, thereby allowing space to display additional text descriptions in the remaining portion of the limited area. Positioning a pointer over the maximum portion of the text causes the complete portion of the text to be displayed. The complete portion of the text can serve as a link to content related to the text. Space is also optimized by providing a thin scroll bar to add space for displaying content within a window.

A method for incremental recognition of ideographic handwriting comprises in order the steps of: (1) entering in a natural stroke order at least one stroke of an ideographic character from a computer entry tablet; (2) providing the at least one stroke to an incremental character recognizer, which produces a hypothesis list of at least one candidate character; (3) displaying a hypothesis list of candidate characters containing the at least one stroke; (4) selecting a correct character from among the candidate characters on the hypothesis list if it a correct character appears thereon; (5) entering in natural stroke order at least one additional stroke of the ideographic character from the computer entry tablet if no candidate character is a correct character; (6) providing the additional stroke(s) to the incremental character recognizer, which produces an updated hypothesis list; (7) displaying the updated hypothesis list of candidate characters containing every stroke; (8) selecting a correct character from among the candidate characters on the updated hypothesis list if it a correct character appears thereon; and (9) repeating steps (5) through (8) until a correct character is selected from the updated hypothesis list.

The invention is a computer implemented method that allows a user to search a text for specific words (e.g. words found in a user-defined library), visually identify any such words found in the text, and have at their fingertips a utility for further exploration of each identified word. The utility, sometimes herein called an ultralink utility, allows a wide variety of options for further exploration of the identified word. The user thereby saves time in explorations; in addition the utility can assist in directing the user to inquiries which may be important for the user.

An audible based electronic challenge system is used to control access to a computing resource by using a test to identify an origin of a voice. The test is based on analyzing a spoken utterance using optimized challenge items selected for their discrimination capability to determine if it was articulated by an unauthorized human or a text to speech (TTS) system.