New here. I am working for a small business on the East Coast. They are into medical transcription. We are looking into implementing a speech recognition system using either pocketsphinx or sphinx4. I tried looking into some of the details and I could get pocketsphinx working with the default models but still learning. My questions are:
Does either pocketsphinx or sphinx4 work out of the box for an use case scenario like medical dictations?
How to go about implementing medical dictation system using CMU sphinx?
Are there any major changes that need to be done to the default setup or config?
Any other relevant points to keep in mind?
Would much appreciate any helpful insights.
Darien
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hi:
New here. I am working for a small business on the East Coast. They are into medical transcription. We are looking into implementing a speech recognition system using either pocketsphinx or sphinx4. I tried looking into some of the details and I could get pocketsphinx working with the default models but still learning. My questions are:
Would much appreciate any helpful insights.
Darien
You should be using Kaldi, it is much more accurate
For medical dictation you have to adapt the dictionary and language model to include medical terms.
For more details see the tutorial
http://cmusphinx.sourceforge.net/wiki/tutorialconcepts
http://cmusphinx.sourceforge.net/wiki/tutoriallmadvanced