A Direct Criterion Minimization based fMLLR via Gradient Descend

Result code in IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F49777513%3A23520%2F13%3A43920629" target="_blank" >RIV/49777513:23520/13:43920629 - isvavai.cz</a>
Result on the web
<a href="http://link.springer.com/chapter/10.1007%2F978-3-642-40585-3_8" target="_blank" >http://link.springer.com/chapter/10.1007%2F978-3-642-40585-3_8</a>
DOI - Digital Object Identifier
<a href="http://dx.doi.org/10.1007/978-3-642-40585-3_8" target="_blank" >10.1007/978-3-642-40585-3_8</a>

Result language
angličtina
Original language name
A Direct Criterion Minimization based fMLLR via Gradient Descend
Original language description
Adaptation techniques are necessary in automatic speech recognizers to improve a recognition accuracy. Linear Transformation methods (MLLR or fMLLR) are the most favorite in the case of limited available data. The fMLLR is the feature-space transformation. This is the advantage with contrast to MLLR that transforms the entire acoustic model. The classical fMLLR estimation involves maximization of the likelihood criterion based on individual Gaussian components statistic.We proposed an approach which takes into account the overall likelihood of a HMM state. It estimates the transformation to optimize the ML criterion of HMM directly using gradient descent algorithm.
Czech name
—
Czech description
—

Project
<a href="/en/project/TA01030476" target="_blank" >TA01030476: Intelligent technologies for improving air traffic security</a><br>
Continuities
P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Publication year
2013
Confidentiality
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Similar results(10)