All of which continues to be perfectly irrelevant as far as voice-morphing goes. Changing the pitch of someone's voice doesn't make them sound like someone else, it just makes them sound like they're on a different pitch. In other words, this entire discussion is largely useless.
Er, yes. Exactly. As we've been mentioning to Malcolm for a while now....
Malcolm, you like to accuse others of obfuscating and squirming away, but what is your purpose in making such a big deal about pitch changing technology? You get so bogged down in details that have absolutely bugger-all to do with the point you apparently started to make, I can't decide whether you just have a pathological fixation on irrelevancies, or are deliberately dragging the discussion away from the point.
Remember what you started by trying to assert? You've probably forgotten, but it was this. That it was possible, in 2001, using publicly-available technology, to fake someone's speaking voice so convincingly that his mother could be convinced she was having a real-time telephone conversation with her son.
That's what you have to demonstrate. Not any irrelevant and silly detail you might have got wrong on the way.
You launched into your argument by referring to musical technology that was capable of altering the pitch of a sung note. However, your choice of examples was pretty unfortunate. Your first attempt was to assert that the original karaoke machines in the 1970s had this capability, which was absolute nonsense, but you wasted an inordinate amount of time trying to prove otherwise.
You wandered past a few kiddies' toys which I suspect even you could see were in the chocolate teapot category, then latched on to this StudioMaster gadget. Which again, unfortunately, didn't actually do what you wanted it to do. Rather than try to find something that did, you again wasted an inordinate amount of time trying to defend your misapprehension that a plug for user-friendly graphic equaliser controls meant that the machine could read the singer's mind!
I've been wondering just how long this was going to go on, but heck, let's try a short-cut.
Psssst, Malcolm. I'll let you into a secret. Electronic instruments capable of changing the pitch of a singer's voice do exist. It's just that the StudioMaster doesn't happen to be one of them. I'm just gobsmacked you haven't been able to track one down.
And yes, the technology is about ten years old, so some sort of instrument of that nature would have been available in 2001. It just wasn't available in a useful form in the 1970s, and it most certainly wasn't incorporated into a karaoke machine of that period.
However, so what? Where does that get you?
First, these things are primarily editing tools. The idea is that you record the voice as it really is, then sit down and fiddle. For one thing, you probably won't be happy with your first try and will want to revise it. And for another, you want a pure sound, and the instruments and background noise will interfere a lot with that.
Could they be used live? Maybe, the electronic response is fast enough. However, problems with ambient noise, and the possibility of the instrument latching on to the wrong sound and trying to "pitch" that, make it unlikely to be terribly satisfactory. Has anything like that yet been incorporated into a karaoke machine? No, though I'm sure someone will give it a try some time. There's never any shortage of takers for a bad idea.
And finally, they're about singing. Not speaking. They're not trying to make one person sound like another, they're about making one person sound a bit better.
Speech recognition is a whole other ball game, and even if you did find the web sites advertising the most sophisticated pitch correction software on the market (no link, find it yourself), you'd have advanced your argument barely a millimetre. You're still several parsecs from actually getting there.
What is it about the sound of someone's voice that allows you to recognise them? It's extraordinarily subtle. The brain isn't algorithmic, it's a pattern-recognition device, and the hearing system is sensitive to the most minute nuances of pattern.
I mentioned my mother was a professional singer. She has spent the last several decades listening to all the best singers of the generations, mainly through Radio 3. She can listen to only a few notes of a song, and tell you who is singing it. And it has nothing to do with pitch. Give her five recordings of the same song, in the same key, and she'll tell you who is singing each one. It's positively uncanny. She can be caught out with recordings of very new artists, but she's particularly rock-solid with people she knew and worked with personally. There's no sort of electronics can turn Richard Lewis into Peter Pears, to the ear of someone who knew them both.
But spoken voices are what we're really talking about, and what's more, spoken voices in a situation where the speaker isn't seen.
I remember idly listening to the car radio one evening, half-hearing a worthy debate among doctors about the ethics of euthanasia. Suddenly my ear was caught and before I even knew why, the image of the face of a close colleague of mine flashed into my mind. Why? Because the producer had decided to get a vet's perspective on the subject, and had brought my colleague into the discussion. Same sort of voice, male, middle-aged, upper-middle-class, all talking about the same thing, but the minute Bob's voice came in, I "saw" his face.
Another example. A weird scene in an episode of
Babylon 5, where the director had disguised an actress so that she was visually unrecognisable. She said just four words. ("You are the hand.") Much discussion about who this mystery woman actually was. My godson was adamant. I was unsure. "Stop looking at her face," he said. "Play the tape again, but don't look at the screen." I did as I was told. Immediately I heard the voice, the image of the actress as she usually looked flashed into my mind. Without the distraction of the disguise, voice recognition systems were spot on.
This is what would have to be achieved in order to fake phone calls. Not anything so mundane as pitch shifting, but the whole subtle pattern of the voice which is a combination of so many things, only starting with pitch, local accent, choice of vocabulary and so on. And not a caricature either. Not Rory Bremner doing Tony Blair, give us a break. Change Peter Pears into Richard Lewis so you fool my mother, just see if you can! (And those two voices are about as similar as you're likely to find.)
And do it live, real-time. Because there's no way you can pre-record stock phrases and be sure you'll have one for every eventuality in the conversation (and remember, Malcolm, Rothschilde doesn't approve plans that might go wrong....). And do it with the right choice of words and mannerisims so that the mother is convinced it's her son.
With a system that was publicly available in 2001.
No way. 8den's point about Hollywood film companies still "looping" inaudible dialogue even in 2007 proves that, even if nothing else does.
This idea is as far into fantasy-land as the idea that a plastic surgeon can sculpt the extra's flesh so that she looks like Marilyn Monroe - and not just on film, under the makeup, but up close and personal to her nearest and dearest.
Maybe one day both will be possible. But not today. And not on 11th September 2001.
Get over it, Malcolm.
Rolfe.