Help achieving real-time auto-tune effect

Question

A while back I played a video game with a character who spoke entirely in auto-tune (Warning: link is NNSFW: Not necessarily safe for work). I thought it was a really fun idea and have been trying to replicate it in real life with varying degrees of success.

I tried various programs and apps like Tune-Me or I-am-T-Pain, but nearly all of them require the user to record an audio snippet, process the audio then play it back. This batch processing approach takes far too long and is cumbersome considering how much auto-tuning I plan to do.

My best attempt so far used an android app called Voloco to produce auto-tune without needing to process the audio in batches. It works well, but there's a very noticable delay (~.5 seconds). In practice this makes it nearly unusable because my voice isn't synced with the auto-tuned output. Anyone nearby can hear my spoken voice separate from the auto-tune and it comes out like 2 people talking over each other (or rather one person singing over another).

My plan is to have a battery-powered speaker hooked up to my phone which is hooked up to microphone. I will have the phone somewhere away from my mouth (maybe in my pocket), the microphone next to my mouth or pinned on my shirt collar and the speaker(s) either mounted on my shoulders or in a backpack. I've got everything working so far, but I need to mute myself from people who could be standing as close as 5 ft from me without preventing the microphone from picking up on my voice.

• This is my plan, but I welcome any advice or alternatives. If someone has a way to achieve real-time auto-tune without all this I'm happy to take a look.

• I want to use to speak to people close to me (5ft - 15ft), not auditoriums or across yards.

• I need to be able to carry the entire setup on my person. I'm willing to use a backpack if necessary, but I'd much prefer a solution that could be stored entirely in my pocket(s)

• Even though it will be disorienting, I don't mind hearing my own spoken voice as well as the auto-tune. I really only care about the people around me.

• My only other restrictions are that I don't want to spend a lot of money (<50$) and that I'm not really capable of doing any tricky electronics. I can may be solder something without electrocuting myself, but passed that I'd need to get a professional.

If there are any other questions leave a comment and I'll update the list. Thanks in advance.

Answer 1

I think the most compact way of doing something like this would be using a vocoder setting on a Korg Kaoss Pad. I'm not an expert on Kaoss Pads but I'm sure they have a vocoder patch. This takes one signal and modulates it with another, so you should be able to modulate the pitch of your voice with a tune either programmed in to the Kaoss Pad or from something like an external MP3 player. You would also need a microphone and some form of amplification, but it should be possible in much less space than a backpack. A good music shop would be able to tell you how get it working. You might also want to look up the term "vocoder" in your mobile device app store. Autotune is a brand name and is more of a studio thing than a real time performance thing. Autotune is really about changing the pitch to be right on the note. Mostly, it is meant to be unnoticeable, but Taylor Swift and countless other modern pop stars aren't fooling me. I think you might be looking for general vocal effect processing.

As for isolating your microphone, that really comes down to the type and quality of microphone and the environment you are using it in.

Answer 2

Perhaps a handheld massager placed somewhere near the back/chest might give the desired effect.

^{Example: http://www.homedics.com/percussion-action-handheld-massager.html}

Answer 3

Place a spinning fan near your mouth and talk through it - it can give a interesting autotuned type voice if done correctly.

Answer 4

The original voice methods that gave rise to auto-tune involved playing sound from an instrument into either the throat or mouth. The famous "talking guitar" is the oldest of these I'm aware of; it used a headphone type speaker with a tube to feed the sound into the performer's mouth as he lip-synced or whispered the words.

This is low-tech stuff, originally operated from a guitar and tube amp, though it would probably work for any sufficiently loud instrument in the human vocal range. The performer can open his throat (breathing in or out) to add resonance, as long as the sound is within the range he'd normally sing. Using a microphone/PA to amplify the resulting voice effect will make it loud enough to fill a room.