Modeling Duration via Lattice Rescoring

It is often acknowledged that HMMs do not properly model phone and word durations. In this paper phone and word duration models are used to improve the accuracy of state-of-the-art large vocabulary speech recognition systems. The duration information is integrated into the systems in a rescoring of word lattices that include phone-level segmentations. Experimental results are given for a conversational telephone speech (CTS) task in French and for the TC-Star EPPS transcription task in Spanish and English. An absolute word error rate reduction of about 0.5% is observed for the CTS task, and smaller but consistent gains are observed for the EPPS task in both languages.