Options for Modelling Temporal Statistical Dependencies in an Acoustic Model for ASR

In this paper we consider the combination of hidden Markov models based on Gaussian mixture densities (GMM-HMM) and linear dynamic models (LDM) as the acoustic model for automatic speech recognition systems. In doing so, the individual strengths of both models, i.e. the modelling of long-term temporal dependencies by the GMM-HMM and the direct modelling of statistical dependencies between consecutive feature vectors by the LDM, are exploited. Phone classification experiments conducted on the TIMIT database indicate the prospective use of this approach in continuous speech recognition.