Speech recognize and voice control a.k.a. JARVIS

Hy!

First of all, sorry about my bad English, I hope you could understand me .
I have a plan. I make my own intelligence house, with voice/speech recognition and control. And I hope the RasPi will answer, and comment that like JARVIS from the Ironman movie.
I don't have Pi now, but the supplier told me I will get it in this week.
First, I want to add some USB microphone, and I really hope it will work perfect and fast. Then I hope I successful install this python program: http://code.google.com/p/pyspeech/. Ok, I know, it's a windows based program and I know it impossible to connect it with RasPi, but it's amazing answer for my questions. So it's just a dream .
Then I find this: http://mikepultz.com/2011/03/accessing- ... nt-page-2/ It's use the google own server for the recognize. The problem with this solution, I want to record the speech and send it to the server.
If this will work, I post the recognized strings to the cleverbot.com and say the answer. There's my work is over, because I will have JARVIS.
It's configure platform, because I have the speech in string, and I control with this everything in the house (via GPIO).
My biggest problem is I don't find any perfect voice and speech recognizer openSource or (L)GPL licensed program.

Someone could offer some good speech recognize program for RasPi?

Try pocketsphinx from here:
http://cmusphinx.sourceforge.net/

The setup is a bit technical, but it is open source, and has been designed to work on machines with limited resources (like smartphones) so could be a good fit for the rpi. I've been meaning to try building apps with it for a while...

I've managed to get speech recognition working somewhat: http://t.co/g5d3pDLF

I've also posted instructions and source code on my website: http://www.aonsquared.co.uk/raspi_voice_control

I have no idea why it's less accurate running on the raspberry pi than on my laptop, using the exact same acoustic models! Any help on this would be much appreciated.

I've managed to get speech recognition working somewhat: http://t.co/g5d3pDLF

I've also posted instructions and source code on my website: http://www.aonsquared.co.uk/raspi_voice_control

I have no idea why it's less accurate running on the raspberry pi than on my laptop, using the exact same acoustic models! Any help on this would be much appreciated.

You should modify it so that you have to address it as "Computer" before it accepts commands, kinda like batman I have a droid so I just wish just one of those siri ripoffs would have that "feature"

Definately cool and something that I want to have a look at. Just want my master thesis to end. Voice control XMBC together with my Arduino would be awesome.

aonsquared: It's awesome. Amazing. Thanx for your link. I hope I got the RasPi tomorrow, so I'll try this as soon as possible.

Try simon listens _http://simon-listens.org/index.php?id=122&L=1, you can write your own commands, responses and other things. I have played with it and it was pretty simple to have a server and multiple clients...phones were to be added in the future release.

I would suggest trying google's api. It is much better than pocketsphinx.

I did a voice control thing. You can find it here:
http://stevenhickson.blogspot.com/2013/ ... ry-pi.html
Here is a video demo:
http://www.youtube.com/watch?v=6NHklmXMouY