A complete overview of distant automatic speech recognition The performance of
conventional Automatic Speech Recognition (ASR) systems degrades dramatically as soon …

We analyze a simple hierarchical architecture consisting of two multilayer perceptron (MLP)
classifiers in tandem to estimate the phonetic class conditional probabilities. In this …

We describe the development of a speech recognition system for conversational telephone
speech (CTS) that incorporates acoustic features estimated by multilayer perceptrons (MLP) …

Automatic speech recognition depends on large amounts of transcribed speech recordings
in order to estimate the parameters of the acoustic model. Recording such large speech …

Abstract Incorporating long-term (500-1000 ms) temporal information using multi-layered
perceptrons (MLPs) has improved performance on ASR tasks, especially when used to …

It is well known that human speech processing capabilities far surpass the capabilities of
current automatic speech recognition and related technologies, despite very intensive …

Provided is a method and apparatus for transforming a speech feature vector. The method
includes extracting a feature vector required for speech recognition from a speech signal …

MLP based front-ends have shown significant complementary properties to conventional
spectral features. As part of the DARPA GALE program, different MLP features were …

This thesis is about speaker-adaptive vowel identification by humans and by machines. In
this first chapter we introduce the research topics and our approaches. We start with a short …

This chapter gives a quick overview on the current speech recognition technology. Front-end
processing, that is, preprocessing and feature extraction, is emphasized, since it is here …