METHOD OF CREATING AN ACOUSTIC MODEL FOR A SPEECH RECOGNITION SYSTEM

摘要

To create an acoustic model (U) for a speaker (3) as the user of a voice recognition system (6), a predetermined number of different group models that are typical of speaker groups (27, 28) are generated as initial models (Gi) and a weighted linear combination of the initial models (Gi) is created using previously determined weight factors (gi) that are specific to the acoustic model (U).