The book is a good read too.
The title :"How to Wreck a Nice Beach" is odd isn't it?
It's actually a vocoder distortion of the phrase "how to recognise speech".
The book follows the history of the vocoder from its invention by Homer Dudley (Bell Labs) in 1928 and its use in war, to its contribution to music today.
The word Vocoder is a combination of two words.
Voice + Code = Vocoder
Dudley invented it to reduce the cost and improve the distance of transmitting vocal signals across the Atlantic. Back in those days transatlantic copper cables were very expensive and distances were huge. Vocoded signals were essentially a form of data compression for voice........ like MP3 is for music today.
I have a few vocoders .. the Korg VC10, a EMS 1000 and the Roland VP-03. These are great, though if you really want to know how these things work you can't go past a modular. There aren't however many modular vocoders around these days. Doepfer used to produce one in their Eurorack A-100 series.
Sadly these are hard to find today.
These came under the code A-129. There were 5 modules in the series, though the 1st two (A-129/1 & A-129/2) are I think the most essential.
If you were back in 1928, you could imagine someone speaking into the A-129/1 in NYC and the A-129/2 being at the listening station in London.
The speech input is first chopped up, analysed & converted into 15 control voltage outputs by this VOC A-129/1 Vocoder Analysis Section module.
The A-129/1 uses 15 steeply sloping filters (13 bandpass, 1 LP & 1 HP). Attached to each filter is an envelope follower which produces the associated control voltage.
These control voltages are transmitted across the transatlantic cable to the A-129/2 module in London.
Here the 15 control voltages are decoded
The A129/2 basically has another 15 filters, but associated with each filter is a VCA.
The filter/VCA's job is to reconstruct the original voice.
The cool thing about this modular vocoder is that these 15 CVs can be modified between the 2 modules.
..... attenuators, slew limiters, LFOs, inverters, etc etc could be used to get interesting results.
The "instrument input" is where you would plug the "carrier" signal .... use something like a VCO.
For best results, the original voice & the carrier signal need to have a similar frequency spectra. ........something like a sawtooth is best as it has a dense audio spectrum. "A square wave has only half as many harmonics, and triangle and sinewaves are completely unsuitable" (A129 manual).
During WWII vocoding was used to encript voice transmissions. The control voltages could be sent over radio, The receiving unit had to have the same filter configuration to decode the messages correctly.
In 1968, Robert Moog developed one of the first solid-state musical vocoders for the electronic music studio of the University at Buffalo.
In 1979 , Moog Music released this vocoder
Below is a pic of Kraftwerks vocoder... made around early 1970.
Buchla 296
It's a lovely module.
In euro, Verbos has a similar module. The Bark Filter processor.
Like the 296, you will need two in order to perform vocoder effects.
+ Doepfer A-129 Vocoder Subsystem
No comments:
Post a Comment