hi i am planning to do an robot controlled by voice..can any one help how do i start with..its enough if i can able to control with some 5 or 6 voice commands..pls suggest me..
I’ve built a simple one:
I’ve built a simple one: https://www.robotshop.com/letsmakerobots/node/2490
But it uses a software on PC for voice recognition. If you are looking for a voice recognition chip, it may be quite expansive (e.g. the HM2007).
If your robot is big enough, a PC software base solution is the cheapest and best approach. Or you may look for a sphinx port to 32bit embedded system.
I don’t have much idea on a voice recognition chip solution.
thank you for your
thank you for your suggestion… let i search something regarding dis…can u say how to do with an pc for voice recognition alone? is it simple??
Please see my answer in my
Please see my answer in my voice robot thread.
In general it is not simple but doable in a couple of days. If you use Microsoft speech SDK, it may be even faster.
You can do it with the VRbot
You can do it with the VRbot module. I intend to use that one or something similar for my robot soon.
http://www.robotshop.us/vrbot-voice-recognition-module-for-robonova-and-other-robots.html
hi
hi…i m tring to make a voice controled robot…can anyone tell me how to do the speech recognition throughh SDK
** check this out, i used**
check this out, i used EasyVR, not optimal, but workable…
http://www.youtube.com/watch?v=2FUMGUOvU-g&feature=player_embedded
Speech Recognition Software
You should take a look at BitVoicer (http://www.bitsophia.com/BitVoicer.aspx). It does much more than just 5 or 6 voice commands, it is cheap and easy to implement.
Thanks for the info! I wish
Thanks for the info!
I wish i know earlier!
EasyVR is quite crap in terms of coding, command training, and operating in noisy environment!
I think I might probably work with speech recognition API in windows for now, which is free :-p
if that doesn’t work out, i might try what you posted! 
thans again!
Voice Recognition
Find yourself a old laptop that still has a parallel port on it and make sure it has a mic plug in on it.
You can used win95 or win98 to do this if you are on a budget.
I used VB6 and programmed mine, it recognizes about 20 commands.
If you need help with getting started with the code let me know i can put some samples up for you to work from.
MyRobotLab
The backend is Sphinx but there are other pluggable Speech-To-Text engines available, such as GoogleSTT …
You can find an easy example in an example script - called talkBack.py - which says the words recognized back.
With a few modifications Gael got it to move around his InMoov robot (many many servos)
http://www.youtube.com/watch?feature=player_embedded&v=W62Wfz1xqYg
Still a work in progress, but it allows the ability to plug in a new VR if available.
Impressive
The design and the amount of detail that has been put into Gael’s robot is very impressive.
When the video shows the robot doing the moves it gets kind of erie as it is just a torso and makes me think of one of those bodies from the walking dead.
I have had similiar issues with voice recognition, it seems that one syllable words work the best for voice recognition.
The other thing I have found out about voice recognition and the words used is that if two or more words are relatively close together in their sound such and “One” and “On” the commands can be confused and while when we speak a word as “Record” the software in most cases understands it as “Rey-Cord” so the pronounciation of words has to be considered.
What I have found that does help is to use a blue tooth head set like the ones made for gammers or a voice activated microphone so that the background noise is canceled out.