Kernel approximation methods for speech recognition

MaRDI QIDQ5381112zbMATH OpenFDO

Authors

Publication date 7 June 2019

Full work available at URL https://arxiv.org/abs/1701.03577, http://jmlr.csail.mit.edu/papers/v20/17-026.html

zbMATH Keywords

acoustic modelingautomatic speech recognitiondeep neural networksfeature selectionkernel methodslogistic regression

Mathematics Subject Classification ID

Generalized linear models (logistic models) (62J12) Artificial neural networks and deep learning (68T07) Pattern recognition, speech recognition (68T10)

Recommendations

• Speech recognition by composition of weighted finite automata
• Rational kernels: theory and algorithms
• A kernel trick for sequences applied to text-independent speaker verification systems
• Continuous speech recognition by convolutional neural networks
• Kernel methods in machine learning

Cites work

• 10.1162/1532443041827925
• A simple proof of Stirling's formula for the gamma function
• Approximation by superpositions of a sigmoidal function
• Core vector machines: fast SVM training on very large data sets
• Coresets, sparse greedy approximation, and the Frank-Wolfe algorithm
• Deep learning
• Efficient online and batch learning using forward backward splitting
• Finding approximate local minima faster than gradient descent
• scientific article; zbMATH DE number 6378127 (Why is no real title available?)
• Kernel approximation methods for speech recognition
• Multilayer feedforward networks are universal approximators
• Nonparametric and semiparametric models.
• Overall risk criterion estimation of hidden Markov model parameters
• The Application of Hidden Markov Models in Speech Recognition
• Training invariant support vector machines
• Universal kernels

Cited in

(9)

• Scalable high-resolution forecasting of sparse spatiotemporal events with kernel methods: a winning solution to the NIJ ``Real-time crime forecasting challenge
• Continuous speech recognition by convolutional neural networks
• Non-parametric policy search with limited information loss
• Study on continuous speech recognition based on bottleneck features for Lhasa-Tibetan dialect
• scientific article; zbMATH DE number 1414403 (Why is no real title available?)
• Support Vector Classifier with Linguistic Interpretation of the Kernel Matrix in Speaker Verification
• Kernel-Based Feature Extraction with a Speech Technology Application
• Kernel approximation methods for speech recognition
• Numerical schemes for signature kernels

Describes a project that uses

Uses Software

• AlexNet
• ImageNet
• word2vec
• DARPA TIMIT
• COFFIN

This page was built for publication: Kernel approximation methods for speech recognition

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5381112)