Abstract: A method for segmenting a compound word in an unrestricted natural-language input is disclosed. The method comprises receiving a natural-language input consisting of a plurality of characters. Next, a set of probabilistic breakpoints based on a probabilistic breakpoint analysis is constructed in the natural-language input. A plurality of linkable components is identified by traversal of substrings of the natural-language input delimited by the set of probabilistic breakpoints. Finally, a segmented string consisting of a plurality of linkable components spanning the natural-language input is returned. The segmented string can be interpreted as a compound word.

Abstract: A method for processing structured text is provided. Tokenized text is created from structured text in accordance with a predetermined set of tokenizer rules set forth in a token pattern knowledge base (13), in which each tokenizer rule defines a simplex constituent. Parsed text is created from tokenized text in accordance with a predetermined set of parser rules set forth in a parser rule knowledge base (15), in which each parser rule defines a complex constituent. Processed text is created from parsed text in accordance with a predetermined set of interpreter rules set forth in an interpretation knowledge base (17), in which each interpreter rule defines a message element corresponding to a simplex or complex constituent, whereby the processed text identifies and provides an interpretation of the message elements of the corresponding structured text for a useful purpose, such as text-to-speech synthesis.

Abstract: A method for segmenting a compound word in an unrestricted natural-language input is disclosed. The method comprises receiving a natural-language input consisting of a plurality of characters. Next, a set of probabilistic breakpoints based on a probabilistic breakpoint analysis is constructed in the natural-language input. A plurality of linkable components is identified by traversal of substrings of the natural-language input delimited by the set of probabilistic breakpoints. Finally, a segmented string consisting of a plurality of linkable components spanning the natural-language input is returned. The segmented string can be interpreted as a compound word.

Abstract: A method (300), device (408), and system (400) provide part-of-speech disambiguation for words based on hybrid neural-network and stochastic processing. The method disambiguates the part-of-speech tags of text tokens by obtaining a set of probabilistically annotated tags for each text token, determining a locally predicted tag for each text token based on the local context of the text token, determining an alternative tag for each text token based on the expanded context of the text token, and choosing between the locally predicted tag and the alternative tag when the locally predicted tag and the alternative tag are different.

Abstract: A method (700), device (1101), and system (1100) provide generalized bidirectional island-driven chart parsing based on congruency checking to prevent edge overgeneration for robust and efficient parsing of a word graph. The method prevents edge overgeneration by selecting, in accordance with a predetermined scheme, a candidate edge with a starting vertex, an ending vertex, a label, and a congruence key for entry in a chart from an agenda of edges, selecting an edge equivalence set in the chart that matches the starting vertex, the ending vertex, and the label of the candidate edge, and entering the candidate edge into the chart if the congruence key of the candidate edge fails to match the congruence key of any edge in the edge equivalence set.

Abstract: A method (400, 900, 1000, 1100), device (1205, 1306) and system (1207, 1308) provide memory-efficient encoding of the random-access lexicon for a text-to-speech synthesis system. The method generates word pronunciations based on efficient retrieval of stored pronunciation information and disambiguation information for an input word, by dividing the input word into a stem and a maximal suffix having a maximal suffix code and then generating at least one word pronunciation based on the stem and the maximal suffix.