This paper describes the Arabic broadcast transcription system fielded by IBM in the GALE Phase 3.5 machine translation evaluation. Key advances compared to our Phase 2.5 system ...
George Saon, Hagen Soltau, Upendra Chaudhari, Step...
A novel frame-wise model adaptation approach for reverberationrobust distant-talking speech recognition is proposed. It adjusts the means of static cepstral features to capture th...
L1 (also referred to as the 1-norm or Lasso) penalty based formulations have been shown to be effective in problem domains when noisy features are present. However, the L1 penalty...
In this paper we evaluate a method for generating synthetic speech at high speaking rates based on the interpolation of hidden semi-Markov models (HSMMs) trained on speech data re...
Michael Pucher, Dietmar Schabus, Junichi Yamagishi
Improved acoustic modeling can significantly decrease the error rate in large-vocabulary speech recognition. Our approach to the problem is twofold. We first propose a scheme that...