Hi,
So, I like to play around with speech synthesizer. There's a particular speech synthesizer called DECtalk, which by the way is a upgraded version of the speech synthesizer Stephen Hawking used. DECtalk can, among other things, be programmed to sing, so I've programmed the Queen song, '39.
I transposed the song to g major insteat of g sharp major. For the instrumentation I used a midi-file from the internet, played it through my Yamaha-keyboard, and added some extra drums.
The link is: link
I would love to hear your thoughts about it.
Best regards:
Aksel Christoffersen
Wow, that's crazy, what an interesting exercise. How many years will it be before we won't be able to tell the difference between a synthesized voice and a real one? Not too long, I'm sure.
Yes, it was very fun.
DECtalk is a synthesizer from the 80's and 90's, so it's a bit outdated, however it's the best synthesizer with a singing capability, I think.
There already exist various synthesizers, that it's almost impossible to tell the difference with a real human voice. These synthesizers are made by recording lots and lots of human speech, and then a program cuts it into bits to be used in synthesized sentences.
akse0435, it's absolutely great, man!!!
Since i've downloaded the song, went on to hear it twice a day and it's never make me boring!
Can you made DecTalk to sing Man on the Prowl, for example? Not asking about urgency, just any time you'll be able.
In fact, i assume this synthesizer can make good result with any Queen song which is in the folk key (or sung drawlingly).
In theory it would be possible to recreate a specific human voice through synthesis. Of course a human voice is very complex, but this is technique is already done with instruments. Some of which are very convincing.
Imagine that 10 years in the future you can buy Freddie in a Box, and Elvis in a Box