Learn More
We describe a new approach to speech recognition, in which all Hidden Markov Model (HMM) states share the same Gaussian Mixture Model (GMM) structure with the same number of Gaussians in each state. The model is defined by vectors associated with each state with a dimension of, say, 50, together with a global mapping from this vector space to the space of(More)
We describe an acoustic modeling approach in which all phonetic states share a common Gaussian Mixture Model structure, and the means and mixture weights vary in a subspace of the total parameter space. We call this a Subspace Gaussian Mixture Model (SGMM). Globally shared parameters define the subspace. This style of acoustic model allows for a much more(More)
Although research has previously been done on multilingual speech recognition, it has been found to be very difficult to improve over separately trained systems. The usual approach has been to use some kind of “universal phone set” that covers multiple languages. We report experiments on a different approach to multilingual speech recognition,(More)
Preparation of a lexicon for speech recognition systems can be a significant effort in languages where the written form is not exactly phonetic. On the other hand, in languages where the written form is quite phonetic, some common words are often mispronounced. In this paper, we use a combination of lexicon learning techniques to explore whether a lexicon(More)
In this paper we present a novel approach for estimating feature-space maximum likelihood linear regression (fMLLR) transforms for full-covariance Gaussian models by directly maximizing the likelihood function by repeated line search in the direction of the gradient. We do this in a pre-transformed parameter space such that an approximation to the expected(More)
PURPOSE The aims were to determine if the maximum standardized uptake value (SUV(max)) of the primary tumor as determined by preoperative (18)F-fluoro-2-deoxyglucose ((18)F-FDG) positron emission tomography (PET) is an independent predictor of overall survival and to assess its prognostic value after stratification according to pathological staging. (More)
BACKGROUND Controversy exists regarding the association of cigarette smoking and renal dysfunction, particularly among African Americans, who are disproportionately affected by chronic kidney disease; therefore, we evaluated the relationship between cigarette smoking and rapid renal function (RRF) decline in the Jackson Heart Study. METHODS AND RESULTS(More)
Genetic alterations in the genes expressing drug metabolizing enzymes can make an individual susceptible to various cancers. This study detects the polymorphisms at CYP1A1, GSTM1, and GSTT1 genes in a section of North Indian population and determines the susceptibility to oral submucous fibrosis (OSF). In this case–control study one hundred and two OSF(More)
The rapid growth of information systems has resulted in new requirements of data indexing and retrieval. Several indexing techniques have been developed, with inverted lists, B+ trees and hashing being most extensively used. We propose a new hash indexing technique for faster search time by using a hash table and resolving collision using a bit-tuple tree.(More)