Abstract: Systems and methods for predictive conversion of user text input in one language or character set to another language or character set are disclosed. The method may generally include converting a partial user input in a non-target language or character set to a converted partial user input in a target language or character set and computing likely possible converted completions of the converted partial user input based on a likely completions module. The converting may be based on a dictionary of non-target language or character set to target language or character set mappings. The likely completions module may be a database or a model trained by performing statistical analysis on a database of user inputs or a subset thereof. The subset of user inputs may be selected from those from a particular geographic location or time period and/or from the user or the user's current session.

Abstract: A method and an apparatus for performing spoken language translation are provided, wherein a speech input is received comprising at least one source language. The speech input comprises words, sentences, and phrases in a natural spoken language. Source expressions are recognized in the source language. Misrecognitions of the source expressions resulting from factors comprising noise and speaker variation are minimized by the generation of intermediate data structures that encode at least one recognition hypothesis. Furthermore, misrecognitions are minimized by the generation of candidate recognized source expressions by processing the intermediate data structures using models comprising a general language model and a domain model. A recognized source expression is selected and confirmed by a user through a user interface. The recognized source expressions are translated from the source language to a target language, and a speech output is synthesized from the translated target language source expressions.

Abstract: At least one speech input is received and at least one token is generated from speech input. Morphemes of the tokens are reduced to at least one feature. Furthermore, an inflection type of the token is identified. At least one dictionary is searched for entries comprising features that match the features reduced from the morphemes. At least one lexical feature structure is generated for the token by inserting at least one morphological feature associated with the inflection type into the entry feature. An output is provided comprising at least one lexical feature structure.

Abstract: A method and an apparatus for style control in natural language recognition and generation are provided, wherein an acoustic input is received comprising at least one source language. The acoustic input comprises words, sentences, and phrases in a natural spoken language. Source expressions are recognized in the source language. Style parameters are determined for the source expression. The style parameters may be extracted from the source expression, set by the user, or randomly selected by the natural language system. A recognized source expression is selected and confirmed by a user through a user interface. The recognized source expressions are translated from the source language to a target language. An acoustic output is generated from the translated target language source expressions using the style parameters. The style parameters comprise variations selected from a group comprising formality, local dialect, gender, and age variations.

Abstract: A method and an apparatus for performing spoken language translation are provided, wherein a speech input is received comprising at least one source language. The speech input comprises words, sentences, and phrases in a natural spoken language. Source expressions are recognized in the source language. Misrecognitions of the source expressions resulting from factors comprising noise and speaker variation are minimized by the generation of intermediate data structures that encode at least one recognition hypothesis. Furthermore, misrecognitions are minimized by the generation of candidate recognized source expressions by processing the intermediate data structures using models comprising a general language model and a domain model. A recognized source expression is selected and confirmed by a user through a user interface. The recognized source expressions are translated from the source language to a target language, and a speech output is synthesized from the translated target language source expressions.

Abstract: In a language translation system, a method for providing a guide for a user. In one embodiment, the method comprises the steps of receiving an input that is representative of at least one word in a source language, and generating at least one recognition hypothesis in the source language in response to the input. The method further comprise the steps of selecting a best hypothesis from the at least one recognition hypothesis in the source language, presenting the best hypothesis in the source language to a user, and presenting alternatives to a portion of the best hypothesis in the source language to the user. The method further includes receiving an indication of a choice of one of the alternatives from the user, and presenting a revised version of the best hypothesis including the alternative chosen to the user.

Abstract: A method and apparatus for adaptive speech recognition hypothesis are provided, wherein a number of ordered recognition hypotheses are generated and presented in response to a received speech input comprising natural spoken language. Generation of the recognition hypotheses comprises assigning basic probabilities to at least one basic component of the speech input using language models and calculating an overall probability of each of the recognition hypotheses using the assigned basic probabilities. The best hypothesis is selected by a user from the recognition hypotheses. Hypothesis generation is adapted in response to the selected best hypotheses, wherein the selected hypothesis is analyzed, a list comprising the basic components of the selected best hypothesis and the assigned basic probabilities is generated, credit is assigned to the basic components of the selected hypothesis by raising the assigned basic probabilities, and the basic probabilities of the language model are renormalized.

Abstract: In a portable unit, a method for performing spoken language translation. The method includes the steps of receiving at least one speech input comprising at least one source language, and recognizing at least one source expression of the at least one source language. The method also includes the steps of translating the recognized at least one source expression from the at least one source language to at least one target language, synthesizing at least one speech output from the translated at least one target language, and providing the at least one speech output. A set of source language expressions and a set of target language expressions are remotely extensible in some embodiments, using various communication methods. One embodiment further comprises the step of minimizing misrecognitions of the at least one source expression, wherein the misrecognitions result from factors comprising noise and speaker variation.

Abstract: Syntactic analysis is performed on an input and on entries of a bilingual example database using at least one parse tree. The parse trees, formed using a context-free grammar, comprise a number of nodes and each node comprises at least one production rule. Furthermore, at least one node comprises nested production rules. The nested production rules comprise production rules for different combinations of the linguistic constituents of the input. The syntactic analysis comprises recognizing linguistic constituents, ordering the linguistic constituents, representing the linguistic constituents using an adapted feature structure analysis representation, and manipulating the adapted feature structure analysis representation using a natural language parser. The syntactic analysis further comprises generalizing surface variations in the input and the entries of the example database in order to increase the translation efficiency.

Abstract: A method and apparatus for parsing in a spoken language translation system are provided, wherein an input is received comprising at least one input sentence or expression. A parsing table is accessed and consulted for a next action, wherein the parser looks up in the next action in the parsing table. During parsing operations, the parser may perform shift actions and reduce actions. In performing a shift action, a next item of the input string is shifted onto a stack or intermediate data structure of the parser. A new parse node is generated, and a feature structure or lexical feature structure of the shifted input item is obtained from a morphological analyzer and associated with the new parse node. The new node is placed on the stack or intermediate data structure. In performing a reduce action, a grammar rule and an associated compiled feature structure manipulation are applied.

Abstract: The present invention is a system of integrated, computer-based processes for monolingual information development. and multilingual translation. An interactive text editor enforces lexical and grammatical constraints on a natural language subset used by the authors to create their text, which they help disambiguate to ensure translatability. The resulting translatable source language text undergoes machine translation into any one of a set of target languages, without the translated text requiring any postediting.

Abstract: A computer-based method and system for monolingual document development, which includes the steps of entering into a text editor input text in a source language, checking the input text against vocabulary source language constraints, and providing interactive feedback relating to the input text if non-constrained source language is present. The method and system also includes the steps of checking for syntactic grammatical errors and semantic ambiguities in the source text, and providing interactive feedback to remove syntactic grammatical errors and semantic ambiguities in the source text.

Abstract: The present invention is a system of integrated, computer-based processes for monolingual information development and multilingual translation. An interactive text editor enforces lexical and grammatical constraints on a natural language subset used by the authors to create their text, which they help disambiguate to ensure translatability. The resulting translatable source language text undergoes machine translation into any one of a set of target languages, without the translated text requiring any postediting.