Hewlett-packard Development Company, L.P.

Digital Equipment Corporation

Compaq Information Technologies Group, L.p.

Pedro J Moreno

Beth Teresa Logan

Jean-Manuel Van Thong

David Goddeau

Brian Scott Eberman

Purdy Ho

Simon Kasif

David M Blei

Edward Whittaker

Baris Suzek

There is provided a method for classifying, identifying or verifying an object by representing the object by a respective sequence of vectors, modeling the sequence of vectors with a respective generative model such that the object is represented by the generative model, computing the distances between the generative models to form one or many kernel matrices based on the distance metric, and using the kernel matrices to classify, identify or verify the object. There is provided a system for determining a classification of an object having a representation module for representing an object by a respective sequence of vectors, a modeling module for modeling the sequence of vectors with a respective generative model such that the object is represented by the generative model, a distance computing module for calculating the distances between the generative models to form one or many kernel matrices based on the distance metric, and a determination module for classifying, identifying or verifying the object based on the kernel matrices.

Method and apparatus for object identification, classification or verification

A speech recognition system provides a subword decoder and a dictionary lookup to process a spoken input. In a first stage of processing, the subword decoder decodes the speech input based on subword units or particles and identifies hypothesized subword sequences using a particle dictionary and particle language model, but independently of a word dictionary or word vocabulary. Further stages of processing involve a particle to word graph expander and a word decoder. The particle to word graph expander expands the subword representation produced by the subword decoder into a word graph of word candidates using a word dictionary. The word decoder uses the word dictionary and a word language model to determine a best sequence of word candidates from the word graph that is most likely to match the words of the spoken input.

Vocabulary independent speech recognition system and method using subword units

Computer method and apparatus of analyzing ECG signals of a subject include receiving a subject electrocardiogram signal and comparing it against signal patterns of known cardiac syndromes. A library of example predefined signals is employed. Distance measures indicating similarity of the subject signal to the example predefined signals are produced and form a sequence of vectors. The sequence of vectors are input into a classifier which determines existence of signal patterns indicative of any cardiac syndromes in the subject.

Cardiac diagnostic system and method

An electronic document searching system or word searching system which when given an input, expands the input as a function of acoustic similarity and/or word sequence occurrence frequency. Results of the system are alternative input words or phrases. The alternative input words or phrases are output from the system for further processing.

Method to expand inputs for word or document searching

Large amounts of multimedia data are transmitted over information networks in the form of a digital stream, analog video, or text captioning. Often, repetitions such as paid advertisements, theme music at the commencement of a TV broadcast, and common jingles and slogans occur in these streams. Detection of repetitions in a transmitted signal such as streaming audio or video is described, and includes extracting a plurality of samples from the information stream and accumulating the samples into segments comprising an interval of the transmitted signal. A vector indicative of the samples in each of the segments is generated, and each of the vectors in the segments is correlated to generate a covariance matrix, or signature, corresponding to the segment. Each of the covariance matrices are aggregated into a sequence of covariance matrices and compared to other covariance matrices to generate a distance matrix. The distance matrix includes a distance value, indicative of the similarity between the distance matrices, as a result of the comparing of each matrix. The distance matrix is then traversed to determine similar sequences of covariance matrices.

System and method for detecting repetitions in a multimedia stream

Computer method and apparatus for reading and analyzing ECG signals includes applying a plurality of heart condition detectors to a subject ECG signal. Each detector produces a respective indication of likelihood of certain heart conditions existing in the subject. A lattice having annotations of the different detector heart conditions is formed from the detector indications. The lattice enables medical personnel to navigate through and hence more easily read the ECG signal data. The lattice effectively provides an indexed or annotated version of the subject ECG signal.

Computer method and system for reading and analyzing ECG signals

A method and apparatus transforms typically differing length text string representations (i.e., sequences) of biological fragments into uniform length representations. A comparison database stores a predefined number of known biological sequences. A comparison routine compares and scores a subject sequence against each known sequence in the database. Each individual score (one for each known sequence in the database) serves as a vector element forming a fixed length vector representation of the subject sequence. Vector length equals the predefined number of known biological sequences in the database. Scoring is a probability or an occurrence count of the known biological sequence in the subject sequence.

Computer method and apparatus for uniform representation of genome sequences

Computer method and apparatus for segmenting text streams is disclosed. Given is an input text stream formed of a series of words. A probability member provides working probabilities that a group of words is of a topic selected from a plurality of predetermined topics. The probability member accounts for relationships between words. A processing module receives the input text stream and using the probability member determines probability of certain words in the input text stream being of a same topic. As such, the processing module segments the input text stream into single topic groupings of words, where each grouping is of a respective single topic.

Computer method and apparatus for segmenting text streams

A method and apparatus are provided for refining time alignments of closed captions. The method automatically aligns closed caption data with associated audio data such that the closed caption data can be more precisely indexed to a requested keyword by a search engine. Further, with such a structure, the closed captions can be made to appear and disappear on a display screen in direct relation to the associated spoken words and phrases. Accordingly, hearing impaired viewers can more easily understand the program that is being displayed.

Method for refining time alignments of closed captions

In a computerized method for processing speech signals, first vectors representing clean speech signals are stored in a vector codebook. Second vectors are determined from dirty speech signals. Noise and distortion parameters are estimated from the second vectors. Third vectors are predicated, based on estimated noise and distortion parameters. The third vectors are used to correct the first vectors. The third vectors can then be applied to the second vectors to produce corrected vectors. The corrected vectors and the first vectors can be compared to identify first vectors which resemble the corrected vectors.

Growing community of inventors

Cambridge, MA, United States of America

Pedro J Moreno