Automatic Speech Recognition for Amharic

In this presentation I have discussed the major components - speech corpus, acoustic model, pronunciation dictionary and language model - of the Amharic Automatic Speech Recognition System (ASRS). Recognition results of different Amharic ASRSs have also been presented. Finally I have pointed out the conclusions that have been drawn from our research and the recommendations that are to be considered for the development and use of Amharic ASRS.

The full presentation is attached here.