Articulatory features for "meeting" speech recognition

“Meeting” speech, for example from the RT-04S task, contains a mixture of different speaking styles that leads to word error rates higher than 25% even when close-talking microphones are being used. The problem is even more serious, as word error rates are particularly high when speakers use a clear speaking mode, for example because they want to stress an important point. Previous work showed that an approach that combines standard phonebased acoustic models with models detecting the presence or absence of “Articulatory Features” such as “Rounded” or “Voiced” can improve ASR performance particularly for these cases. This paper presents a discriminative approach to automatically computing from training or adaptation data the feature stream weights needed for the above approach, therefore presenting a framework for integrating articulatory features into existing automatic speech recognition systems. We find a 7% relative improvements on top of our best RT-04S system using discriminative adaptation.