I'd like to create application for mobile phone which would be responsible for speech recognition. I thought about creating MIDlet for phone with Symbian and PocketSphinx.
However, is it possible that I can somehow access the ability of mobile phone to support "voice dialing". In fact speech recognition for mobile phones is already implemented. But I don't know whether it is possible to have access to this speech recognition by voice dialing.
If I need to use PocketSphinx, can anybody suggest me good (-> htt p:/ /cm usphinx.sourceforge.net/sphinx4/doc/Sphinx4-faq.html#j2me) [delete space-bars in the link] and cheap mobile phone, which can enable me speech recognition?
In general I'd like to enable communication: human <-> mobile phone <-> server.
User would run application on mobile phone and create some data based on speech recognition and then exit the application, forcing it to send this obtained data to the server. I guess POST method of httpconnection on MIDlet would be enough to send from mobile phone and Servlet created with Eclipse for TomCat to receive this little text file (how to save to file and read from file in the case of MIDlet? - I mean in source code).
The crucial thing is this speech recognition. It can be done in one of the following ways:
1. using already implemented voice dialing, if it is possible to communicate with it :-), is it?
2. using open-source project like PocketSphinx
On the other hand, there are some services when somebody can call special number and it talks to something. It says "received messages", then "yes", this kind of things. Can you give me, please, any links to tutorials how to create my own menu based on speech recognition of this kind?
By the way, can you suggest me good phone for my purpose? It can't be more expensive than 230 american dollars. It is to be bought in Central Europe.