Pocketsphinx is pretty far behind the state of the art in speech recognition algorithms. It also doesn't have a very large acoustic model. You can greatly impr

Pocketsphinx is pretty far behind the state of the art in speech recognition algorithms. It also doesn't have a very large acoustic model. You can greatly improve the recognition by using a restricted language model, however.