Abstract: A system and an article of manufacture for discriminative learning via hierarchical transformations, which includes obtaining a model of a first set of data, two or more data transformations, and a second set of data, evaluating the two or more data transformations to determine which data transformation will most effectively modify the second set of data to match the model, and selecting the data transformation that will most effectively modify the second set of data to match the model based on the evaluation.

Abstract: Systems and methods for applying feature-space maximum likelihood linear regression (fMLLR) to correlated features are provided. A method for applying fMLLR to correlated features, comprises mapping the correlated features into an uncorrelated feature space, applying fMLLR in the uncorrelated feature space to obtain fMLLR transformed features, and mapping the fMLLR transformed features back to a correlated feature space.

Abstract: Systems and methods for applying feature-space maximum likelihood linear regression (fMLLR) to correlated features are provided. A method for applying fMLLR to correlated features, comprises mapping the correlated features into an uncorrelated feature space, applying fMLLR in the uncorrelated feature space to obtain fMLLR transformed features, and mapping the fMLLR transformed features back to a correlated feature space.

Abstract: Systems and methods for training networks are provided. A method for training networks comprises receiving an input from each of a plurality of neural networks differing from each other in at least one of architecture, input modality, and feature type, connecting the plurality of neural networks through a common output layer, or through one or more common hidden layers and a common output layer to result in a joint network, and training the joint network.

Abstract: A method includes providing a deep neural network acoustic model, receiving audio data including one or more utterances of a speaker, extracting a plurality of speech recognition features from the one or more utterances of the speaker, creating a speaker identity vector for the speaker based on the extracted speech recognition features, and adapting the deep neural network acoustic model for automatic speech recognition using the extracted speech recognition features and the speaker identity vector.

Abstract: A system and an article of manufacture for discriminative learning via hierarchical transformations, which includes obtaining a model of a first set of data, two or more data transformations, and a second set of data, evaluating the two or more data transformations to determine which data transformation will most effectively modify the second set of data to match the model, and selecting the data transformation that will most effectively modify the second set of data to match the model based on the evaluation.

Abstract: Techniques for discriminative learning via hierarchical transformations. A method includes obtaining a model of a first set of data, two or more data transformations, and a second set of data, evaluating the two or more data transformations to determine which data transformation will most effectively modify the second set of data to match the model, and selecting the data transformation that will most effectively modify the second set of data to match the model based on the evaluation.

Abstract: In connection with speech recognition, the design of a linear transformation ??p×n, of rank p×n, which projects the features of a classifier x?n onto y=?x?p such as to achieve minimum Bayes error (or probability of misclassification). Two avenues are explored: the first is to maximize the ?-average divergence between the class densities and the second is to minimize the union Bhattacharyya bound in the range of ?. While both approaches yield similar performance in practice, they outperform standard linear discriminant analysis features and show a 10% relative improvement in the word error rate over known cepstral features on a large vocabulary telephony speech recognition task.

Abstract: In a speech recognition system, the combination of a log-linear model with a multitude of speech features is provided to recognize unknown speech utterances. The speech recognition system models the posterior probability of linguistic units relevant to speech recognition using a log-linear model. The posterior model captures the probability of the linguistic unit given the observed speech features and the parameters of the posterior model. The posterior model may be determined using the probability of the word sequence hypotheses given a multitude of speech features. Log-linear models are used with features derived from sparse or incomplete data. The speech features that are utilized may include asynchronous, overlapping, and statistically non-independent speech features. Not all features used in training need to appear in testing/recognition.

Abstract: In a speech recognition system, the combination of a log-linear model with a multitude of speech features is provided to recognize unknown speech utterances. The speech recognition system models the posterior probability of linguistic units relevant to speech recognition using a log-linear model. The posterior model captures the probability of the linguistic unit given the observed speech features and the parameters of the posterior model. The posterior model may be determined using the probability of the word sequence hypotheses given a multitude of speech features. Log-linear models are used with features derived from sparse or incomplete data. The speech features that are utilized may include asynchronous, overlapping, and statistically non-independent speech features. Not all features used in training need to appear in testing/recognition.

Abstract: Methods and arrangements using lattice-based information for unsupervised speaker adaptation. By performing adaptation against a word lattice, correct models are more likely to be used in estimating a transform. Further, a particular type of lattice proposed herein enables the use of a natural confidence measure given by the posterior occupancy probability of a state, that is, the statistics of a particular state will be updated with the current frame only if the a posteriori probability of the state at that particular time is greater than a predetermined threshold.

Abstract: In a speech recognition system, the combination of a log-linear model with a multitude of speech features is provided to recognize unknown speech utterances. The speech recognition system models the posterior probability of linguistic units relevant to speech recognition using a log-linear model. The posterior model captures the probability of the linguistic unit given the observed speech features and the parameters of the posterior model. The posterior model may be determined using the probability of the word sequence hypotheses given a multitude of speech features. Log-linear models are used with features derived from sparse or incomplete data. The speech features that are utilized may include asynchronous, overlapping, and statistically non-independent speech features. Not all features used in training need to appear in testing/recognition.