Author Options:

How plausible would the idea of making a voice-controlled chess set be? Answered

Yesterday, after playing a round in a chess tournament, I thought of the idea of a voice-controlled chess set. Mac has done a voice-controlled chess app, so why not make the same thing, but with a real set? My idea is to have each piece have a board with a railroad-esque track system for the pieces to move around on in response to a voice control unit also incorporated into the set. My main concern was how to circumvent the voice-control mechanism being unable to determine which individual piece to move if more than one of that piece is on the board and capable of the move being made (for instance, if the player said knight to E5, and both knights are capable of moving there, which one does it move?), but then decided that the voice activation could be activated by stating a square name, and the command affects the piece on that square. I know I would have to scale the set up a bit to fit the electronics, but is this idea plausible? How much would the technology to make it cost? 


An arduino-powered chess-playing robot was created here by mJusticz last year.

I see no reason why that project cannot be modified to accept voice commands.

Going by the costs mJusticz posted, I would estimate the cost of your idea to be $400 or higher, depending on what materials, tools and skills you already possess.

mJusticz is still active here, so I would suggest contacting him directly.

Take a look at THIS as a brilliant implementation and build log of a chess playing robot.  The chess playing algorithm is programmed into the Arduino he's using for control.  Once you've built that, all you need to do is to interface one of THESE Sparkfun voice recognition modules to it and off you jolly well go.
We'll expect an Instructable by next week.

Yes, it can be done - All the information and parts are available -  but it would require very good mechanics and programming skills and a good few months of labour.

A way to simplify the mechanical aspect would be to buy a ready-made robot arm to do the moving, but again you'd have to integrate this with a chess-playing algorithm and voice recognition module.

Best way to differentiate ambiguous moves is to specify start-point and end-point, e.g. insttead of 'knight to e5' you'd say 'd3 to e5'.  This would make programming marginally easier too.

this souds possible, expensive, super cool. if you finish it post it on insturctables. Good Luck

Voice operation, as anyone who has tried it, will tell you is not very successful unless the conditions are just right.

You can test this on your PC quite easily as all Win ops have this facility.

The mechanics would be fairly costly to build and even more so If you want it to look professional rather than engineered.

Cinema robotics and special effects may as well be the same thing!!

Pick and place arms would work, Magnets under the table on an x,Y gantry might work, Overhead gantry and an electro magnet could work., Making each playing piece mobile with small wheels and a power supply should work.

It's a great idea with potentially a big budget and a lot of time invested. Great talking point though!

An on screen version may be somewhat easier to engineer.

You'd have to program the movements similar to any other electronic chess game and each piece would have to have a specific name. So for a knight, you'd have knight 1 and knight 2 to allow the user to select the appropriate chess piece and position to move on the board.

The biggest challenge I believe would be their actual movements. Using a knight as another example, they move in a specific way and would likely encounter a number of obstacles, being other chess pieces. So you'd have to find a way around that (pun intended).

It is a very interesting idea. Post an instructable if you can build it. :)

I imagine a skeleton hand picking up the piece and putting it down again in the new spot - no obstacles (well, aside from building the arm and making it work :-)

If you want to seriously do this, you'd have to map out the whole process and consider all the possible obstacles along the way. Not only the physical ones, but the practical ones as well. I think the main reason that electronic chess games don't utilize pieces is due to the problems you're having to overcome. I think its possible, and could be a very cool idea. If the large claw concept doesn't seem feasible (because pieces can get accidentally knocked over), then consider having each piece programmed to move according to voice commands; or have the voice command and movement built into the board, and sensors attached to the bottom of each piece. The movements could be achieved with magnets on the pieces and on the mechanism within the board. That might be more difficult, but you'll figure out the best method by going through the options based on your tools and programming knowledge.

Interesting idea, though, how would you handle capturing pieces, or castling?

Really cool idea, also a realy hard idea to create.

Good luck