Uncertainty driven Compensation of Multi-Stream MLP Acoustic Models for Robust ASR.

Ramón Fernandez Astudillo Alberto Abad João Paulo Neto

Published in: INTERSPEECH (2012)

Keyphrases

multi stream
automatic speech recognition
speech recognition
hidden markov models
audio visual speech recognition
neural network
noisy environments
acoustic models
broadcast news
high level
audio visual
data points
image classification
conditional probabilities
speech signal