Hi,
In my Application user will see a picture and he should say name of the item and the recognizer should check if user said correct name or not.
i want to know is there anyway to implement this application for persian ( farsi ) language for about fixes 200 words/pictures?
Thanks in Advance
Last edit: arash 2016-04-20
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I recommend you to train model with large vocabulary, that will save you a lot of work when your vocabulary will change, if you add one more word or something like this. You can include the target words into training database, but it is better to also add other phrases, not just target 200 words.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hi,
In my Application user will see a picture and he should say name of the item and the recognizer should check if user said correct name or not.
i want to know is there anyway to implement this application for persian ( farsi ) language for about fixes 200 words/pictures?
Thanks in Advance
Last edit: arash 2016-04-20
Sure, there is a way. You build a model for Farsi as described in our tutorial http://cmusphinx.sourceforge.net/wiki/tutorialam and then implement the application as described in other part of tutorial http://cmusphinx.sourceforge.net/wiki/tutorialandroid and it should work. You can use keyword spotting mode to check for each word.
wow , fast answer! thanks.just a very basic question, is it enough to train the model with this fixed(200 words) or not?
I recommend you to train model with large vocabulary, that will save you a lot of work when your vocabulary will change, if you add one more word or something like this. You can include the target words into training database, but it is better to also add other phrases, not just target 200 words.