Dec 19, 2006 · In this paper, we study discriminative training of acoustic models for speech recognition under two criteria: maximum mutual information ...
Abstract—In this paper we study discriminative training of acoustic models for speech recognition under two criteria: max- imum mutual information (MMI) and ...
Abstract—In this paper, we study discriminative training of acoustic models for speech recognition under two criteria: max-.
In this paper, we study discriminative training of acoustic models for speech recognition under two criteria: maximum mutual information (MMI) and a novel ...
Bibliographic details on Discriminative Estimation of Subspace Constrained Gaussian Mixture Models for Speech Recognition.
Discriminatively trained full-covariance Gaussian mixture models have been shown to outperform its corresponding diagonal-covariance models on large ...
ABSTRACT. We introduce speech recognition based on subspace adaptation of a shared Gaussian Mixture Model (GMM). This refers to adapta-.
A standard approach to automatic speech recognition uses hidden Markov models whose state dependent distributions are Gaussian mixture models.
Missing: Estimation | Show results with:Estimation
Abstract—A standard approach to automatic speech recog- nition uses Hidden Markov Models whose state dependent distributions are Gaussian mixture models.
People also ask
What is the Gaussian mixture model for speech recognition?
Is the Gaussian mixture model generative or discriminative?
What is the subspace Gaussian mixture model?
Oct 18, 2013 · Discriminatively trained full-covariance Gaussian mixture models have been shown to outperform its corresponding diagonal-covariance models ...
Missing: Estimation | Show results with:Estimation