I'm using CMU Sphinx for some months now and I'm really enjoying it. It is so useful for me. I had had some difficulties but I found the solution for each one here in this forum and now I have another problem. I'm creating a robot for my school project and I'm improving the accuracy day by day. It works really fine. My problem now, and i hope the last one, is the OOV words. I need a way to avoid OOV words necessarily. Please show me an efficient way. Thanks for everything!
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I'm using CMU Sphinx for some months now and I'm really enjoying it. It is so useful for me. I had had some difficulties but I found the solution for each one here in this forum and now I have another problem. I'm creating a robot for my school project and I'm improving the accuracy day by day. It works really fine. My problem now, and i hope the last one, is the OOV words. I need a way to avoid OOV words necessarily. Please show me an efficient way. Thanks for everything!
You can add more words to the dicationary, there are no easy ways.
Modern toolkits like wav2letter have lexicon-free speech recognition:
https://github.com/facebookresearch/wav2letter/tree/master/recipes/models/lexicon_free
Can I integrate wav2letter on CMU Sphinx or should I start over again doing all my work with this toolkit?
You have to start over.
Is this toolkit easy to learn or use?
Depends on your skills, overall, not very easy and you will need a GPU card.