Abstract: Aspects of the technology employ a machine translation quality prediction (MTQP) model to refine datasets that are used in training machine translation systems. This includes receiving, by a machine translation quality prediction model, a sentence pair of a source sentence and a translated output (802). Then performing feature extraction on the sentence pair using a set of two or more feature extractors, where each feature extractor generates a corresponding feature vector (804). The corresponding feature vectors from the set of feature extractors are concatenated together (806). And the concatenated feature vectors are applied to a feedforward neural network, in which the feedforward neural network generates a machine translation quality prediction score for the translated output (808).

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, relating to language models stored for digital language processing. In one aspect, a method includes the actions of generating a language model, including: receiving a collection of n-grams from a corpus, each n-gram of the collection having a corresponding first probability of occurring in the corpus, and generating a trie representing the collection of n-grams, the trie being represented using one or more arrays of integers, and compressing an array representation of the trie using block encoding; and using the language model to identify a second probability of a particular string of words occurring.

Abstract: One feature of the present invention uses the parsing capabilities of a structured language model in the information extraction process. During training, the structured language model is first initialized with syntactically annotated training data. The model is then trained by generating parses on semantically annotated training data enforcing annotated constituent boundaries. The syntactic labels in the parse trees generated by the parser are then replaced with joint syntactic and semantic labels. The model is then trained by generating parses on the semantically annotated training data enforcing the semantic tags or labels found in the training data. The trained model can then be used to extract information from test data using the parses generated by the model.

Abstract: The present invention involves using one or more statistical classifiers in order to perform task classification on natural language inputs. In another embodiment, the statistical classifiers can be used in conjunction with a rule-based classifier to perform task classification.

Abstract: Methods are disclosed for estimating language models such that the conditional likelihood of a class given a word string, which is very well correlated with classification accuracy, is maximized. The methods comprise tuning statistical language model parameters jointly for all classes such that a classifier discriminates between the correct class and the incorrect ones for a given training sentence or utterance. Specific embodiments of the present invention pertain to implementation of the rational function growth transform in the context of a discriminative training technique for n-gram classifiers.

Abstract: Systems, methods, and apparatuses, including computer program products, are provided for representing language models. In some implementations, a computer-implemented method is provided. The method includes generating a compact language model including receiving a collection of n-grams from the corpus, each n-gram of the collection having a corresponding first probability of occurring in the corpus and generating a trie representing the collection of n-grams. The method also includes using the language model to identify a second probability of a particular string of words occurring.

Abstract: Systems, methods, and apparatuses, including computer program products, are provided for representing language models. In some implementations, a computer-implemented method is provided. The method includes generating a compact language model including receiving a collection of n-grams from the corpus, each n-gram of the collection having a corresponding first probability of occurring in the corpus and generating a trie representing the collection of n-grams. The method also includes using the language model to identify a second probability of a particular string of words occurring.

Abstract: One feature of the present invention uses the parsing capabilities of a structured language model in the information extraction process. During training, the structured language model is first initialized with syntactically annotated training data. The model is then trained by generating parses on semantically annotated training data enforcing annotated constituent boundaries. The syntactic labels in the parse trees generated by the parser are then replaced with joint syntactic and semantic labels. The model is then trained by generating parses on the semantically annotated training data enforcing the semantic tags or labels found in the training data. The trained model can then be used to extract information from test data using the parses generated by the model.

Abstract: One feature of the present invention uses the parsing capabilities of a structured language model in the information extraction process. During training, the structured language model is first initialized with syntactically annotated training data. The model is then trained by generating parses on semantically annotated training data enforcing annotated constituent boundaries. The syntactic labels in the parse trees generated by the parser are then replaced with joint syntactic and semantic labels. The model is then trained by generating parses on the semantically annotated training data enforcing the semantic tags or labels found in the training data. The trained model can then be used to extract information from test data using the parses generated by the model.

Abstract: A system and method for creating a mnemonics Language Model for use with a speech recognition software application, wherein the method includes generating an n-gram Language Model containing a predefined large body of characters, wherein the n-gram Language Model includes at least one character from the predefined large body of characters, constructing a new Language Model (LM) token for each of the at least one character, extracting pronunciations for each of the at least one character responsive to a predefined pronunciation dictionary to obtain a character pronunciation representation, creating at least one alternative pronunciation for each of the at least one character responsive to the character pronunciation representation to create an alternative pronunciation dictionary and compiling the n-gram Language Model for use with the speech recognition software application, wherein compiling the Language Model is responsive to the new Language Model token and the alternative pronunciation dictionary.

Abstract: A statistical classifier is constructed by estimating Naïve Bayes classifiers such that the conditional likelihood of class given word sequence is maximized. The classifier is constructed using a rational function growth transform implemented for Naïve Bayes classifiers. The estimation method tunes the model parameters jointly for all classes such that the classifier discriminates between the correct class and the incorrect ones for a given training sentence or utterance. Optional parameter smoothing and/or convergence speedup can be used to improve model performance. The classifier can be integrated into a speech utterance classification system or other natural language processing system.

Abstract: A method and apparatus are provided for adapting a language model. The method and apparatus provide supervised class-based adaptation of the language model utilizing in-domain semantic information.

Abstract: A system and method for creating a mnemonics Language Model for use with a speech recognition software application, wherein the method includes generating an n-gram Language Model containing a predefined large body of characters, wherein the n-gram Language Model includes at least one character from the predefined large body of characters, constructing a new Language Model (LM) token for each of the at least one character, extracting pronunciations for each of the at least one character responsive to a predefined pronunciation dictionary to obtain a character pronunciation representation, creating at least one alternative pronunciation for each of the at least one character responsive to the character pronunciation representation to create an alternative pronunciation dictionary and compiling the n-gram Language Model for use with the speech recognition software application, wherein compiling the Language Model is responsive to the new Language Model token and the alternative pronunciation dictionary.

Abstract: Methods are disclosed for estimating language models such that the conditional likelihood of a class given a word string, which is very well correlated with classification accuracy, is maximized. The methods comprise tuning statistical language model parameters jointly for all classes such that a classifier discriminates between the correct class and the incorrect ones for a given training sentence or utterance. Specific embodiments of the present invention pertain to implementation of the rational function growth transform in the context of a discriminative training technique for n-gram classifiers.

Abstract: A system and method for creating a mnemonics Language Model for use with a speech recognition software application, wherein the method includes generating an n-gram Language Model containing a predefined large body of characters, wherein the n-gram Language Model includes at least one character from the predefined large body of characters, constructing a new language Model (LM) token for each of the at least one character, extracting pronunciations for each of the at least one character responsive to a predefined pronunciation dictionary to obtain a character pronunciation representation, creating at least one alternative pronunciation for each of the at least one character responsive to the character pronunciation representation to create an alternative pronunciation dictionary and compiling the n-gram Language Model for use with the speech recognition software application, wherein compiling the Language Model is responsive to the new Language Model token and the alternative pronunciation dictionary.

Abstract: A method and apparatus are provided for storing parameters of a deleted interpolation language model as parameters of a backoff language model. In particular, the parameters of the deleted interpolation language model are stored in the standard ARPA format. Under one embodiment, the deleted interpolation language model parameters are formed using fractional counts.

Abstract: Methods are disclosed for estimating language models such that the conditional likelihood of a class given a word string, which is very well correlated with classification accuracy, is maximized. The methods comprise tuning statistical language model parameters jointly for all classes such that a classifier discriminates between the correct class and the incorrect ones for a given training sentence or utterance. Specific embodiments of the present invention pertain to implementation of the rational function growth transform in the context of a discriminative training technique for n-gram classifiers.

Abstract: A computer-implemented method of indexing a speech lattice for search of audio corresponding to the speech lattice is provided. The method includes identifying at least two speech recognition hypotheses for a word which have time ranges satisfying a criteria. The method further includes merging the at least two speech recognition hypotheses to generate a merged speech recognition hypothesis for the word.

Abstract: A speech segment is indexed by identifying at least two alternative word sequences for the speech segment. For each word in the alternative sequences, information is placed in an entry for the word in the index. Speech units are eliminated from entries in the index based on a comparison of a probability that the word appears in the speech segment and a threshold value.

Abstract: An index for searching spoken documents having speech data and text meta-data is created by obtaining probabilities of occurrence of words and positional information of the words of the speech data and combining it with at least positional information of the words in the text meta-data. A single index can be created because the speech data and the text meta-data are treated the same and considered only different categories.