Phonetic context adaptation for improved speech recognition

摘要

The present invention relates to a computerized method and corresponding means and a computer program product for automatically generating from a first speech recognizer a second speech recognizer, said second speech recognizer being adapted to a certain domain. The first speech recognizer comprising a first acoustic model with a first decision network and its corresponding first phonetic contexts. The current invention suggests using said first acoustic model as a starting point for the adaptation process. A second acoustic model with a second decision network and its corresponding second phonetic contexts for the second speech recognizer is generated by re-estimating the first decision network and said corresponding first phonetic contexts based on domain-specific training data.