S
Sangita Sharma
Researcher at Oregon Health & Science University
Publications - 12
Citations - 1391
Sangita Sharma is an academic researcher from Oregon Health & Science University. The author has contributed to research in topics: Feature (machine learning) & Word error rate. The author has an hindex of 7, co-authored 7 publications receiving 1366 citations. Previous affiliations of Sangita Sharma include International Computer Science Institute.
Papers
More filters
Proceedings ArticleDOI
Tandem connectionist feature extraction for conventional HMM systems
TL;DR: A large improvement in word recognition performance is shown by combining neural-net discriminative feature processing with Gaussian-mixture distribution modeling.
Proceedings ArticleDOI
Temporal patterns (TRAPs) in ASR of noisy speech
Hynek Hermansky,Sangita Sharma +1 more
TL;DR: The proposed neural TRAPs are found to yield significant amount of complementary information to that of the conventional spectral feature based ASR system, which results in improved robustness to several types of additive and convolutive environmental degradations.
Proceedings Article
TRAPS - classifiers of temporal patterns.
Hynek Hermansky,Sangita Sharma +1 more
TL;DR: The work proposes a radically different set of features for ASR where TempoRAl Patterns of spectral energies are used in place of the conventional spectral patterns.
Journal ArticleDOI
Relevance of time-frequency features for phonetic and speaker-channel classification
TL;DR: A large database of hand-labeled fluent speech is used to compute the mutual information between a phonetic classification variable and one spectral feature variable in the time–frequency plane, and the joint mutual information (JMI) between the phonetic Classification variable and two feature variables in thetime-frequency plane.
Proceedings ArticleDOI
Feature extraction using non-linear transformation for robust speech recognition on the Aurora database
TL;DR: It is shown that after a non-linear transformation, a number of features can be effectively used in a HMM-based recognition system.