do you need to write it yourself, or are you looking to use/purchase a third party package that will do it for you?
There are only two hard things in computer science: cache invalidation, naming things, and off-by-one errors
Joined: Oct 25, 2009
We are three student of b.tech. we choose a project on speech recognition to do in this semester.
But after studying a lot of on this project we are not understanding that how can start to write code of this.
So please give some Idea to start.
Or if anyone have code for this.Please give me that code so that at least we will start to write code for this.
I will be very grateful for that.....
Please help me....
We have read this document.
But we have instructed not to use inbuilt java library to make this project like javax.speech.
we can use HMM(Hidden markov model ) for speech recognition.
So if someone have code for speech recognition using HMM or have some idea.Then please help me...
That's a tough problem indeed - it's still actively researched, and not everything about it is yet understood. There's an implementation for javax.speech called Sphinx - you may get some information from the web site, even if you can't use the software.
Unfortunately Internet forums cannot be helpful when your question is something like "I don't have a clue about this topic". You would have more chance of getting an answer if you had a specific question. And as for asking for code, I think you're getting a bit ahead of yourself. Shouldn't you be designing something and then looking for code to do that something? (Assuming you're allowed to do your project by soliciting code from other people, that is.)
arun singhal wrote:Please give me that code so that at least we will start to write code for this.
How about this:
that's usually How I start my projects.
But in all seriousness...You can't expect to ask such broad, open questions like this and get specific answers. The broader your question, the broader the answer.
I'd start by doing research. figure out what needs to be done. Are you recording/translating live input or a pre-recorded file? Will it be local, or do you have to connect to a remote server to get the input? What format does the output have to be?
A lot of that you can work on without ever writing a single line of 'speech verification'.
While you work on that code, you can research the other parts.