ROBOVOX by Kraftwerk

[Pallamanski]

Hi friends - happy new year

I have some questions about Kraftwerks Robovox, the machine they used to generate speech synthesis for their music - I hope we have someone here with knowledge on this. So here goes the question:

1) Did someone ever made an emulator of this? Or are there some application or plugin that resembles it in some way? I use XOXOS Syng2 which is a pretty cool alternative

2) If you listen to Kraftwerk's "The Robots" (from The Mix album) you clearly hear Robovox in action:
http://www.youtube.com/watch?v=9P-JaG2PpyI
However the speech synthesis used by Karl Bartos when he went solo sounds in my ear different. Example here:
http://www.youtube.com/watch?v=Dx9IWUlmt2I
Does anyone know if it's a completely different machine or a modification of the Robovox

3) Does anyone know if the phonemes from Votrax SC-01 or SC-02 is aviable on the net in form of samples? I know Kraftwerk used these on some of their compositions

Thanks in advance for your inputs

Patrick

[Kim Olesen]

erased. Read your post wrong. Thought you meant the version of The Robote from the Man Machine album.

[Pallamanski]

Hi friends - happy new year

I have some questions about Kraftwerks Robovox, the machine they used to generate speech synthesis for their music - I hope we have someone here with knowledge on this. So here goes the question:

1) Did someone ever made an emulator of this? Or are there some application or plugin that resembles it in some way? I use XOXOS Syng2 which is a pretty cool alternative

2) If you listen to Kraftwerk's "The Robots" (from The Mix album) you clearly hear Robovox in action:
http://www.youtube.com/watch?v=9P-JaG2PpyI
However the speech synthesis used by Karl Bartos when he went solo sounds in my ear different. Example here:
http://www.youtube.com/watch?v=Dx9IWUlmt2I
Does anyone know if it's a completely different machine or a modification of the Robovox

3) Does anyone know if the phonemes from Votrax SC-01 or SC-02 is aviable on the net in form of samples? I know Kraftwerk used these on some of their compositions

Thanks in advance for your inputs

Patrick

Hi. I'm sorry but you got it a bit wrong. The voice on The Man Machine album is just a vocoder. The first time they used the robo-vox was on The Mix album. It's all over the album and unmistakable. This is the same technique that Bartos used on the Esperanto album. Kraftwerk may have claimed to have invented it before they actually did. They often did that to intrigue people despite in reality often using just off the shelf products in their music.

Hi Kim

Ah I see. On the album Radioactivity they use a speech synth though. For example in Radioland and Ohm Sweet Ohm. Do you know which?

[Goseba]

This may be of interest

http://kraftwerkfaq.hu/equipment.html

[Gamma-UT]

The book Music Non-Stop is worth looking at for info on what was used when although, as I don't think Kraftwerk ever went into detail about what was used where, there is still some guesswork involved.

A good start might be this: Florian Schneider filed for a patent at the EPO for the Robovox (or something very much like it) at roughly about the same time as The Mix was being recorded: http://www.freepatentsonline.com/EP0396141.pdf

It's based on the Votrax SC-02 speech chip anyway.

If you google for Votrax or Type'n'Talk samples, you should find some that are usable.

Retro-games emulators may be your best bet for a Votrax emulation: I don't know of one in audio plugin form.

[Gamma-UT]

Ah I see. On the album Radioactivity they use a speech synth though. For example in Radioland and Ohm Sweet Ohm. Do you know which?

I think Ohm Sweet Ohm is a vocoder. Uranium is a more likely candidate for the Votrax.

[Kim Olesen]

Voices on Radioactivity is also a vocoder.

[Pallamanski]

Ah I see. On the album Radioactivity they use a speech synth though. For example in Radioland and Ohm Sweet Ohm. Do you know which?

I think Ohm Sweet Ohm is a vocoder. Uranium is a more likely candidate for the Votrax.

Hmm I'm positive it's a speech synth through a vocoder though. Because if you listens to the "S" in sweet, you can clearly hear it's a phoneme

[Pallamanski]

The book Music Non-Stop is worth looking at for info on what was used when although, as I don't think Kraftwerk ever went into detail about what was used where, there is still some guesswork involved.

A good start might be this: Florian Schneider filed for a patent at the EPO for the Robovox (or something very much like it) at roughly about the same time as The Mix was being recorded: http://www.freepatentsonline.com/EP0396141.pdf

It's based on the Votrax SC-02 speech chip anyway.

If you google for Votrax or Type'n'Talk samples, you should find some that are usable.

Retro-games emulators may be your best bet for a Votrax emulation: I don't know of one in audio plugin form.

Gamma, thanks a lot for the info I've been doing a lot of search on the net. Basically I was looking for emulations of retro IBM computers which may have had a SC-01 or SC-02 chip as mandatory. I know Votrax came out with Votalker - a cartridge for C64 but I havent found a ROM for it.

Do you have other info where this chip was featured. It really buggles me that this historic peace of history hasn't been made into somekind of software or emulation. eBay doesn't have any chips for sale anymore

[Goseba]

From the link I put above:
"A military speech synthesizer, based on creating phonemes, was used on the song Radioland"
In which case could have been used on other tracks if correct.

[Goseba]

More links to some info, even has an image of the patent application by Florian:

http://www.technopop-archive.com/data_technology.php

http://www.technopop-archive.com/data_setup.php

[deastman]

I know Votrax came out with Votalker - a cartridge for C64 but I havent found a ROM for it.

Do you have other info where this chip was featured. It really buggles me that this historic peace of history hasn't been made into somekind of software or emulation. eBay doesn't have any chips for sale anymore

I picked up a C64 voice synth cartridge and disk at a local electronics surplus house a few years ago. I'm pretty sure it was Votalker. I never did get around to pulling one of my two C64's out of storage to test it out, but I might just have to do that now... assuming the floppy is still readable.

[xoxos]

fyi syng3 should be available by the end of february, but it will sound more like vocaloid than robot voices.

i've never checked into the robovox patent. electric cafe sounds a lot like the concatenated sample speech synths that were predominant in the 90's like dectalk, which is very similar though the samples are from different people. it's about the same quality as shittalker, if you know that one. for singing, if you're game for it, track down the mbrola packages.

farbrausch's synths afaik use the voice synth in windows, which is in the same region technologically.

[Pallamanski]

fyi syng3 should be available by the end of february, but it will sound more like vocaloid than robot voices.

i've never checked into the robovox patent. electric cafe sounds a lot like the concatenated sample speech synths that were predominant in the 90's like dectalk, which is very similar though the samples are from different people. it's about the same quality as shittalker, if you know that one. for singing, if you're game for it, track down the mbrola packages.

farbrausch's synths afaik use the voice synth in windows, which is in the same region technologically.

Hey Xoxos

Do you have any suggestions on where to look for a Votrax Text to Speech? The dayes they were on eBay seem long gone.

Dont be offended but I am not into the vocaloid thing but I wondered, if you somehow got all the phoneme samples from Votrax SC-01 or SC-02 could you make an alternate Syng with Votrax voices - I think that would be awsome as it would surely come close to Robovox

I also wanted to ask you, do you know Elektron Monomachine. It seems to have a pretty interesting voice synth, one example here:
http://www.youtube.com/watch?v=klQ8beKPspA

[xoxos]

hmm.. seeing votrax vids on youtube.. way primitive. i'm working on a synthetic voice intended to be as hifi as possible. it sounds like vocaloid/a real voice spectrally, but it won't sound like vocaloid in terms of being samples, because the synthesis model allows for continuous modulation between phonemes and intensity. it's nowhere near finished yet so i can't really say what kind of feel to expect, i have yet to experiment with forms of variation and other improvements tat will distinguish it from concatenated samples.

i've only ever heard one hifi source > filter model, which was a private development. i'm not aware of any public non-sample based synthesis models to use as an example.

it's an expandable model, i can add voices to it but the process is complex, say a week or two of work. given the improved quality, i expect some people won't like it because it sounds more like a specific person than low fidelity models do (of course it's still more flexible than vocaloid, where you can't shift the formant or character).

so i am looking for good recordings of people speaking or singing that have pleasant voices but lofi robot voices would waste most of the spectral bands of this model..