TRAINING OF SPEAKER-CLUSTERED ACOUSTIC MODELS FOR USE IN REAL-TIME RECOGNIZERS
Jan Vaněk, Josef V. Psutka, Aleš Pražák, Josef Psutka · 2009
The paper deals with training of speaker-clustered acoustic models. Various training techniques Maximum Likelihood, Discriminative Training and two adaptation based on the MAP and Discriminative MAP were tested in order to minimize an impact of speaker changes to the correct function of the recognizer when a response of the automatic cluster detector is delayed or incorrect. Such situation is very frequent e.g. in online subtitling of TV discussions (Parliament meetings). In our experiments the best cluster-dependent training procedure was discriminative adaptation which provided the best trade-off between recognition results with correct and non-correct cluster detector information.