Jump to content

Midibox Vocoder


tr3tton

Recommended Posts

Yeah! That IS cool!

But ofcourse, I was thinking about a DIY module vocoder.. With only the "external" inputs...

I have seen some old DIY analog beasts, but I was wondering if it was possible to to it either in a software for the MB or alternatively in another module with special software?

Maybe its better/easier to build an analog version as a standalone module? But since I'm not really "hip" and "in the game" when it comes to more modern circuits, I thought I'd ask here!

Link to comment
Share on other sites

thanks guys :)

About building your own vocoder: This would be a lot of work. There are a few schematics for analog vocoders found in the net, but if the result should be good you always need half a square meter of PCBs. The most compact one I´ve found is the kit from PAIA. But it has only 8 bands and is more expensive than e.g. a used MAM VF11.

Going digital is not possible without a suitable DSP. No way to do this in software with a PIC. You could use a "Miss Parker" and programm your own algorithm. But that´s the same DSP that is used in the Alesis Metavox, and I suppose Alesis have quite maxed out what is doable on this DSP.

The question always is: Why would you do this? If you just do it because of the fun and to get experience, then do it. You need some DSP programming experience and a lot of time, but it is possible.

But if you want do build a vocoder that sounds better and/or is cheaper than commercial alternatives, then forget it!

The commercial hardware alternatives I´ve already mentioned. The Metavox used in the phone is really great value. It´s not the best vocoder in the world, but the sound is better than in most multi-FX boxes (I´ve heard vocoders in 90´s Digitech, Zoom and Boss FX units, and the Metavox is way better). And you can probably get it at around 30 Euros on eBay.

The MAM VF11 is another beast. I have one and it´s absolutely great. Fully analog, 11 adjustable bands, unvoiced detection with possibility to feed external unvoiced signal. It sounds really fat and the intelligibility is quite good for "only" 11 bands. It´s going for around 100-150 Euros on eBay.

Both machines have definitely a bang for the buck ratio you can´t beat by DIYing.

Seppoman

P.S. perhaps a moderator could move this thread to Design Concepts or Miscellaneous as it doesn´t have anything to do with MBSID?

Link to comment
Share on other sites

Hi all,

I actually wanted to start at some point to code a Vocoder for the "Miss Parker" (http://www.axoris.be) but never really finalized it. I had some piece of floating-point C code that was doing the job but got stuck with filter scaling and fixed-point issues. I can't remember exactly but I think I was targetting initially 16 bands in total however the scaling issues might well have as consequence a reduction of that number of bands.

I still want to revive that project at some point but as always time is lacking...

Best regards,

Lall

Link to comment
Share on other sites

Hi dj3nk,

If you talk about the Miss Parker, it's "unfortunately" an "almost DIY" project in the sense that some components on the board require professional etching.

I think the problem mainly comes from the DAC used to control the input and output gains. That's the MAX5841 or U12 on the schematic and that one has a veeery small pitch. For the rest, it's do-able I guess even though a professionally made board really helps the soldering due to the solder mask they put on it (risks of shorts while soldering are much less).

You can look on that page: http://www.axoris.be/MissParkerDetails.php. There's normally everything required (Gerber files package) to have a board been made by some PCB makers. I know at least one guy who made his own Miss Parker using a such a PCB maker company in his neighboorhood. I'm not the one who did the layout and manufacturing stuff so I don't know if a lot of support was required from our side but in the end it worked.

Best regards,

Lall

Link to comment
Share on other sites

Vocoders can be monsters!

Analog ones can be very versatile and give a good range of effects, but a big job to make, and do require fairly skilled setting up/tuning to sound good. Need a lot of close tolerance parts for good filters too.

Digital ones eat a lot of CPU, if you want real-time. If you want to be able to do all the fancy patching stuff, then you may have to write your own code. No microcontroller can get into the speed range yet, so it's DSP or, if you fancy something very up-to-date, a custom processor running on an FPGA - ther are a couple of designs out on the web somewhere:

http://www56.homepage.villanova.edu/scott.sawyer/fpga/II_freq_domain.htm this article is interesting.

The frustrating thing about doing research for a musical vocoder is the amount of telecoms stuff you find, since they are the major users of the technique, and the telecom stuff isn't much use for music.

The Elektor analogue design is still out there, and is a good one too - I built two for local musicians back when it was current, A little work could bring the design well up to date. You can download the original PDF's from the magazine site I think, and there are other versins on line too.

Elektor and ETI design here, under 'synths'

http://www.timewind.se/

There are quite a few soft vocoders available, some for free, though all need a PC. CSound has a phase vocoder with which you can 'morph' sounds. Csound has a very steep learning curve, I found, if you are a programmer, then it will be a lot easier.

Just a few thoughts

Mike

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
×
×
  • Create New...