Abstract: A method detects events in an accoustic signal subject to cyclostationary background noise by first segmenting the signal into cycles. Features with a fixed dimension are derived from the cycles, such that the timing of the features is relative to a cycle time. The features are normalized using an estimate of the cyclostationary background noise. Then, after the normalizing, a classifier is applied to the features to detect the events.

Abstract: A method for processing a voice command using a statistical dialog model determines a belief state as a probability distribution over states organized in a hierarchy with a parent-child relationship of nodes representing the states. The belief state includes the hierarchy of state variables defining probabilities of each state to correspond to the voice command and a probability of a state of a child node in the hierarchy is conditioned on a probability of a state of a corresponding parent node. A system action is selected based on the belief state.

Abstract: Text is classified by determining text features from the text, and transforming the text features to topic features. Scores are determined for each topic features using a discriminative topic model. The model includes a classifier that operates on the topic features, wherein the topic features are determined by the transformation from the text features, and the transformation is optimized to maximize the scores of a correct class relative to the scores of incorrect classes. Then, a class label with a highest score is selected for the text. In situations where the classes are organized in a hierarchical structure, the discriminative topic models apply to classes at each level conditioned on previous levels and scores are combined across levels to evaluate the highest scoring class labels.

Abstract: An information system includes a prediction engine for predicting an action based on a set of driving state parameters, and a driving history, and a simulation engine for generating a hypothetical scenario by simulating one or a combination of at least one driving state parameter and at least part of the driving history, such that the prediction engine predicts the action for the hypothetical scenario.

Abstract: The embodiments of the invention provide a method in a navigation system, for predicting travel destinations according to a history of destinations. A model used for the prediction incorporates a database of destinations, which can include favorite, i.e., most probable, destinations for a user. The model also uses a context that can include features such as a current time of day, day of week, current location, current direction, past location, weather, and so on. The model infers the destination and destination categories even when the destination is not known precisely. Specifically, a method predicts destinations during travel, based on feature vectors representing current states of the travel, probabilities of destinations and categories of the destinations using a predictive model representing previous states of the travel. A subset of the destinations and categories of the destinations with highest probabilities are output for user selection.

Abstract: A method determines from an input noisy signal sequences of hidden variables including at least one sequence of hidden variables representing an excitation component of the clean speech signal, at least one sequence of hidden variables representing a filter component of the clean speech signal, and at least one sequence of hidden variables representing the noise signal. The sequences of hidden variables include hidden variables determined as a non-negative linear combination of non-negative basis functions. The determination uses the model of the clean speech signal that includes a non-negative source-filter dynamical system (NSFDS) constraining the hidden variables representing the excitation and the filter components to be statistically dependent over time. The method generates an output signal using a product of corresponding hidden variables representing the excitation and the filter components.

Abstract: A method adapts a user interface of a vehicle navigation system. Based on an input vector representing a current state related to the vehicle, probabilities of actions are predicted to achieve a next state using a predictive model representing previous states. Then, a subset of the actions with highest probabilities that minimize a complexity of interacting with the vehicle navigation system are displayed in the vehicle.

Abstract: A method processes an acoustic signal that is a mixture of a target signal and interfering signals by first enhancing the acoustic signal by a set of enhancement procedures to produce a set of initial enhanced signals. Then, an ensemble learning procedure is applied to the acoustic signal and the set of initial enhancement signals to produce features of the acoustic signal.

Abstract: A system and a method recognize speech including a sequence of words. A set of interpretations of the speech is generated using an acoustic model and a language model, and, for each interpretation, a score representing correctness of an interpretation in representing the sequence of words is determined to produce a set of scores. Next, the set of scores is updated based on a consistency of each interpretation with a constraint determined in response to receiving a word sequence constraint.

Abstract: A method converts source speech to target speech by first mapping the source speech to sparse weights using compressive sensing technique, and the transforming, using transformation parameters, the sparse weights to the target speech.

Abstract: Enhanced speech is produced from a mixed signal including noise and the speech. The noise in the mixed signal is estimated using a vector-Taylor series. The estimated noise is in terms of a minimum mean-squared error. Then, the noise is subtracted from the mixed signal to obtain the enhanced speech.

Abstract: An input signal, in the form of a sequence of feature vectors, is transformed to an output signal by first storing parameters of a model of the input signal in a memory. Using the vectors and the parameters, a sequence of vectors of hidden variables is inferred. There is at least one vector hn of hidden variables hi,n for each feature vector xn, and each hidden variable is nonnegative. The output signal is generated using the feature vectors, the vectors of hidden variables, and the parameters. Each feature vector xn is dependent on at least one of the hidden variables hi,n for the same n. The hidden variables are related according to h i , n = ? j , l ? ? c i , j , l ? ? l , n ? h j , n - 1 , where j and l are summation indices. The parameters include non-negative weights ci,j,l, and ?l,n are independent non-negative random variables.

Abstract: Access is obtained to a large reference acoustic model for automatic speech recognition. The large reference acoustic model has L states modeled by L mixture models, and the large reference acoustic model has N components. A desired number of components Nc, less than N, to be used in a restructured acoustic model derived from the reference acoustic model, is identified. The desired number of components Nc is selected based on a computing environment in which the restructured acoustic model is to be deployed. The restructured acoustic model also has L states. For each given one of the L mixture models in the reference acoustic model, a merge sequence is built which records, for a given cost function, sequential mergers of pairs of the components associated with the given one of the mixture models. A portion of the Nc components is assigned to each of the L states in the restructured acoustic model.

Abstract: Text is classified by determining text features from the text, and transforming the text features to topic features. Scores are determined for each topic features using a discriminative topic model. The model includes a classifier that operates on the topic features, wherein the topic features are determined by the transformation from the text features, and the transformation is optimized to maximize the scores of a correct class relative to the scores of incorrect classes. Then, a class label with a highest score is selected for the text. In situations where the classes are organized in a hierarchical structure, the discriminative topic models apply to classes at each level conditioned on previous levels and scores are combined across levels to evaluate the highest scoring class labels.

Abstract: Text is processed to construct a model of the text. The text has a shared vocabulary. The text is partitioned into sets and subsets of texts. The usage of the shared vocabulary in two or more sets is different, and the topics of two or more subsets are different. A probabilistic model is defined for the text. The probabilistic model considers each word in the text to be a token having a position and a word value, and the usage of the shared vocabulary, topics, subtopics, and word values for each token in the text are represented using distributions of random variables in the probabilistic model, wherein the random variables are discrete. Parameters are estimated for the model corresponding to the vocabulary usages, the word values, the topics, and the subtopics associated with the words.

Abstract: A probe is registered with an object by probing the object with the probe at multiple poses, wherein each pose of the probe includes a location and an orientation. A probability distribution of a current location of the probe is represented by a set of particles, and a probability distribution of a current orientation of the probe is represented by a Gaussian distribution for each particle conditioned on the current location. A set of candidate motions is chosen, and for each candidate motion, an expected uncertainty based on the set of particles is determined. The candidate motion with a least expected uncertainty is selected as a next motion of the probe, the probe is moved according to the next motion, and the set of particles is updated using the next pose of the probe.

Abstract: Enhanced speech is produced from a mixed signal including noise and the speech. The noise in the mixed signal is estimated using a vector-Taylor series. The estimated noise is in terms of a minimum mean-squared error. Then, the noise is subtracted from the mixed signal to obtain the enhanced speech.

Abstract: Access is obtained to a large reference acoustic model for automatic speech recognition. The large reference acoustic model has L states modeled by L mixture models, and the large reference acoustic model has N components. A desired number of components Nc, less than N, to be used in a restructured acoustic model derived from the reference acoustic model, is identified. The desired number of components Nc is selected based on a computing environment in which the restructured acoustic model is to be deployed. The restructured acoustic model also has L states. For each given one of the L mixture models in the reference acoustic model, a merge sequence is built which records, for a given cost function, sequential mergers of pairs of the components associated with the given one of the mixture models. A portion of the Nc components is assigned to each of the L states in the restructured acoustic model.

Abstract: A probe is registered with an object by probing the object with the probe at multiple poses, wherein each pose of the probe includes a location and an orientation. A probability distribution of a current location of the probe is represented by a set of particles, and a probability distribution of a current orientation of the probe is represented by a Gaussian distribution for each particle conditioned on the current location. A set of candidate motions is chosen, and for each candidate motion, an expected uncertainty based on the set of particles is determined. The candidate motion with a least expected uncertainty is selected as a next motion of the probe, the probe is moved according to the next motion, and the set of particles is updated using the next pose of the probe.

Abstract: A system and method facilitating speech detection and/or enhancement utilizing audio/video fusion is provided. The present invention fuses audio and video in a probabilistic generative model that implements cross-model, self-supervised learning, enabling rapid adaptation to audio visual data. The system can learn to detect and enhance speech in noise given only a short (e.g., 30 second) sequence of audio-visual data. In addition, it automatically learns to track the lips as they move around in the video.