Speech recognition

Has anyone ever used speech recognition in their projects before? Is the best way to go using something like this? https://www.sparkfun.com/products/10963

If you have used the EasyVR, just to confirm, it does not need to be connected to a PC in order to operate, correct?

I wanted to use this as an easy way to turn off all the lights in the house when going to sleep. So I was gonna leave this (Arduino+EasyVR) always plugged in and waiting for a voice command like "good night house!" which would then trigger an X10 RF transmitter to turn off all the lights.

Does that sound feasible? Is there a warm-up period after the EasyVR hears a sound? Or is it always ready 24/7 to validate what it's heard?


EasyVR was out of stock, you may take a look on this http://forum.arduino.cc/index.php?topic=118616.0 http://arjo129.github.io/uSpeech/

If it's for permanent installation in a house then perhaps you'd consider a PC-based solution. That would enable much more reliable and robust voice detection. I haven't used the Easy VR myself, but people who have described it as giving quite unreliable voice recognition - not really what you want for something that will be listening to everything you say and turning lights on and off (and who knows what else) in response to any commands it thinks it heard.

I agree with PeterH and, in this case, BitVoicer could be the way to go. I think my post can help: http://forum.arduino.cc/index.php?topic=140765.0