Polaxis Talko Is Back!

Speech synthesis has always intrigued me...  ever since i saw the possibilities of a SpeakJet chip... and these FreQTalk videos

Belgian company Polaxis has been pushing the boundaries of speech synthesis for over a decade, and now they’re bringing back a classic: the Talko.

The original Talko was their very first speech synthesizer, later followed by Emy, which added an OLED screen, SD card support, and a whole new level of control. 

Over the years, Polaxis gathered 10 years of user feedback, collected thousands of LPC files, and experimented with new chip emulations. 

The result? Talko returns — with onboard FX!

This new-generation Talko(2?) stays true to its raw, charming character, but now comes packed with fresh features. 

Based on a Speak & Spell emulation (probably running on an Arduino?), it can speak numbers, alphabets, phonemes, and phrase banks, all while running in real time through built-in effects.

It’s designed to be DIY-friendly and will be available both as a Eurorack module or in a quirky standalone “bricky” format (a nod to @errorinstruments).

🔊 Want to hear it live? Come check it out at Dutch Modular Fest, Saturday September 6th in Rotterdam!

Dit bericht op Instagram bekijken

Een bericht gedeeld door Jean-Luc Deladrière (@polaxis)

Most Wanted VII: A-1?? Speech Module

Time to get one of my earlier segments off the shelf.

Since i first started to write this blog, i have posted quite a few 'Most Wanted' posts about modules i would really like to see being made by Doepfer.

In some cases i even did some predictions and/or wild guesses which proved to be not that far from what has come out of the hands of Doepfer lately.

For example... Take a look for example at my blogpost from December 2010, and compare it with the details on Doepfer's forthcoming A-190-8 MIDI-to-Sync Interface module.

A-1?? Speech module,
announced in the 2008 Doepfer brochure

I really wish i could tell you more about this one too, but i can't...

The A-1?? Speech Module was first mentioned in the 2008 A-100 brochure, as a 'synthetic human voice moduled controlled by CV/Gate or MIDI' 

I often wondered what this module would be like, and I have been waiting for more info about it ever since.
My hopes were that it would be a module with a SpeakJet chip inside, just like FLAME's Talking Synth, a module that has always intrigued me...
The variety of the SpeakJet chip ranges from speech-like sounds and synthetic robot voices (as well as beeps and alarms), to retro-style sci-fi sounds.
Due to the structure of the Speakjet (with its complex sound synthesizer, preset sounds and serial interface) it offers an impressive range of possibilities. The SpeakJet is not sample based as you might think, it has its own sound engine.

Video: Speakjet Chip Demo Mode

" In this video I demonstrate the Speakjet chip in demo mode in which you can hear all the pre-loaded vocal and sounds that the chip outputs in demo mode.
FYI: I used the mic connection instead of the 120ohm speaker that they recommend since you can barley hear the chip(in demo mode w/o an amplifier)!"

Uploaded by sergiolcjr

To be honest, i don't even care what speech-chip the A-1?? will have inside, the thought alone of bringing synthesized speech into a Eurorack modular is a good thought.
In my opinion it is hard enough nowadays to invent original and new truly analog modules, so i would welcome the use of more digital chips and circuits integrated in a CV/Gate-controlled modular environment.

Sadly enough, this module disappeared in all later Doepfer brochures.

I will investigate this further and post more about this module later... i hope... ;-)

A few days after posting this Dieter replied to my blogpost;
" The response to the announcement of the speech module was very poor.
And as there is already a similar module available from another manufacturer (Flame Talking Synth) we decided to stop this development and focus on other new modules." 

Find all my earlier 'Most Wanted' posts HERE

Quick Tip III : Vocoder testing

Ideal for ( Doepfer ) vocoder testing;


Try feeding an audio-book or a recording of your favorite radioshow through your system while tweaking your vocoder settings.

Speech-only programmes are perfect for this purpose.

Short demo:
  In the Morning Vocoder test by NetPierre
* original audioclip from: http://NoAgendaShow.com / http://NoAgendaShots.com