Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

voice controled light dimming

Status
Not open for further replies.

danny009

New Member
hi i was just thinking of a new idea and i thought of a bedroom light controller then i thought of a dimmer but then i thought it would be a pain to build or have to stand up to get a remote or press a button then i thought voice controled because i think lots lol i was just wondering if this could be possible using something like a picaxe and a small ciruit im not sure how we could wire it into the mains 240v so any help will be great thanks.
 
Get your voice recognition working first, then worry about the mains interface - to be honest, it sounds a silly idea! - even commercial voice recognition doesn't impress me greatly! :D
 
Voice recognition is easy:
"Bright-ten".
"Dim".
Bark twice and it gets brighter.
Bark one time and it dims.

It is a pain to press a button???
 
sorry should have said that a bit better on some dimming switches you press a button or turn a little dial to change the lights brightness. hmm to get the voice recognition would you simply use something like a microphone to a input on the picaxe thanks
 
danny009 said:
sorry should have said that a bit better on some dimming switches you press a button or turn a little dial to change the lights brightness. hmm to get the voice recognition would you simply use something like a microphone to a input on the picaxe thanks

I suggest you try looking into the difficulties of voice recognition?, which you aren't going to do with a tiny amount of program space and a very slow BASIC interpreter.
 
You can buy voice recognition modules - just add a microcontroller and a away you go!

It shouldn't be hard to get the micro to control the firing point in the cycle in response to voice commands.
 
Get an old computer with a 733 Pentium III in it, and put M$ windows on it, along with a good flavor of the .net suit (i would suggest VB.NET, or C++, or C.), then download the MS Speech SDK. Simple as pie. Just a few lines of code to get a simple voice recognition program, and then another couple of lines to send data out the parallel port.

Simple. ;)
 
Marks256 said:
Get an old computer with a 733 Pentium III in it, and put M$ windows on it, along with a good flavor of the .net suit (i would suggest VB.NET, or C++, or C.), then download the MS Speech SDK. Simple as pie. Just a few lines of code to get a simple voice recognition program, and then another couple of lines to send data out the parallel port.

But rather expensive to build, and to run - plus how reliable is it?, not very from what I've heard!.
 
Nigel Goodwin said:
But rather expensive to build, and to run - plus how reliable is it?, not very from what I've heard!.

How practical would it be to do it on a PIC? :) And how would this be expensive? Yeah, sure, the computer would be running for a long time, but the code would be free, and i am sure you could get a suitable computer for free at a yard sale.

As for reliability, it isn't very good, but, like you yourself said,
Nigel Goodwin said:
which you aren't going to do with a tiny amount of program space and a very slow BASIC interpreter.
Reliability does go up after the "training" sessions are run. I doubt one could match the quality (or lack there of) of the Speech SDK on a simple PICAXE.
 
Nigel Goodwin said:
- it's still really the realm of science fiction!.

How so? It is 100% do-able. It wouldn't be perfect, but it would still work... That is not the definition of "science fiction", Nigel.
 
Marks256 said:
How so? It is 100% do-able. It wouldn't be perfect, but it would still work... That is not the definition of "science fiction", Nigel.

I would expect such a device to work reliably, with different peoples voices, and NOT make any mistakes! - this is a LONGGGGGGGG way off :p
 
Nigel Goodwin said:
But rather expensive to build, and to run - plus how reliable is it?, not very from what I've heard!.
Easy, subsiture Windows for a more stable operating system and it shouldn't fail.

However I do agree that power consumption is an issue.
 
Are there any GOOD speech SDK for linux? If so, i would be very interested into finally learning some C/C++... (yeah, havn't really started that...)
 
Google has the answer
**broken link removed**
http://www.google.com/search?hl=en&...cd=1&q=open+source+speech+recognition&spell=1

Another advantage of using Linux would be that as you don't need graphics you wouldn't bother with the X window system or a destop so you could get away with less RAM.

Also some BIOSes actually underclock the CPU when it isn't being used to save power. You also might be able to configure it to but the system into sleep mode and only wakes up when the microphone recieves sound.
 
I use X10. You'd think it's not very convenient, but I can keep a remote next to my pillow.
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top