Publication

Using Posterior-Based Features in Template Matching for Speech Recognition

Related publications (34)

Graph Chatbot

Chat with Graph Search

Ask any question about EPFL courses, lectures, exercises, research, news, etc. or try the example questions below.

DISCLAIMER: The Graph Chatbot is not programmed to provide explicit or categorical answers to your questions. Rather, it transforms your questions into API requests that are distributed across the various IT services officially administered by EPFL. Its purpose is solely to collect and recommend relevant references to content that you can explore to help you answer your questions.

Using Pitch as Prior Knowledge in Template-Based Speech Recognition

Hervé Bourlard, Guillermo Aradilla

In a previous paper on speech recognition, we showed that templates can better capture the dynamics of speech signal compared to parametric models such as hidden Markov models. The key point in template matching approaches is finding the most similar templ ...

2006

Boosting word error rates

Samy Bengio, Christos Dimitrakakis

We apply boosting techniques to the problem of word error rate minimisation in speech recognition. This is achieved through a new definition of sample error for boosting and a training procedure for hidden Markov models. For this purpose we define a sample ...

2005

Boosting word error rates

Samy Bengio, Christos Dimitrakakis

2005

Using Pitch as Prior Knowledge in Template-Based Speech Recognition

Hervé Bourlard, Guillermo Aradilla

IDIAP2005

Gradient estimates of return distributions

Samy Bengio, Christos Dimitrakakis

We present a general method for maintaining estimates of the distribution of parameters in arbitrary models. This is then applied to the estimation of probability distributions over actions in value-based reinforcement learning. While this approach is simi ...

2005

Boosting word error rates

Samy Bengio, Christos Dimitrakakis

IDIAP2004

Estimates of Parameter Distributions for Optimal Action Selection

Samy Bengio, Christos Dimitrakakis

We present a general method for maintaining estimates of the distribution of parameters in arbitrary models. This is then applied to the estimation of probability distribution over actions in value-based reinforcement learning. While this approach is simil ...

IDIAP2004

To code or not to code

Michael Christoph Gastpar

It is well known and surprising that the uncoded transmission of an independent and identically distributed Gaussian source across an additive white Gaussian noise channel is optimal: No amount of sophistication in the coding strategy can ever perform bett ...

EPFL2002

Pricing Swaptions within the Affine Framework

Pierre Collin Dufresne

We propose a very fast and accurate algorithm for pricing swaptions when the underlying term structure dynamics are affine. The efficiency of the algorithm stems from the fact that the moments of the underlying asset (i.e., a coupon bond) possess simple cl ...

Columbia Business School2002

Transient analysis of adaptive filters

Ali H. Sayed

This paper develops a framework for the mean-square analysis of adaptive filters with general data and error nonlinearities. The approach relies on energy conservation arguments and is carried out without restrictions on the probability distribution of the ...

IEEE2001