On the performance of ITU-T G.723.1 and AMR-NB codecs for large vocabulary distributed speech recognition in Brazilian Portuguese

In this paper, we present the accuracy for large vocabulary distributed continuous speech recognition systems over ITU-T G.723.1 and AMRNB speech codecs. Experiments were conducted using LPC and LSF-derived speech recognition features, CDHMM acoustic models, triphone units and trigram language models for the Brazilian Portuguese.