I\'m writing a program to transcribe audio using CMU Sphinx. I\'m not happy with the quality and I thought maybe I could find a better model. But I don\'t really understand the
