Acoustic word disambiguation with phonological features in Danish ASR

摘要

Phonological features can indicate word class and we can use word class information to disambiguate both homophones and homographs in automatic speech recognition (ASR). We show Danish stod can be predicted from speech and used to improve ASR. We discover which acoustic features contain the signal of st0d, how to use these features to predict stod and how we can make use of stod and stod-predictive acoustic features to improve overall ASR accuracy and decoding speed. In the process, we discover acoustic features that are novel to the phonetic characterisation of stod.