| Author |
Message |
Pehr

Joined: Aug 14, 2005 Posts: 1307 Location: Björkvik, Sweden
Audio files: 2
|
|
|
Back to top
|
|
 |
mister

Joined: Jun 06, 2006 Posts: 11 Location: Washington DC, USA
Audio files: 6
|
Posted: Fri Jun 09, 2006 6:20 am Post subject:
|
 |
|
| highly liquid has been doing this for a while and he is a bad ass. I consider this bending because you are going beyond the traditional purpose of the machine and the sounds it makes are straight out glitch and noise, just like a bent speak! anyone who loves the chaos of circuit bending and the control of MIDI should definitely check out highly liquid. thanks for bringing up this topic PEHR! |
|
|
Back to top
|
|
 |
mi_dach

Joined: Dec 17, 2005 Posts: 133 Location: Sweden
|
Posted: Fri Jun 09, 2006 7:09 am Post subject:
|
 |
|
In a recent mail to Texas Instruments, they said to me
"we do not keep data sheets for products that have been out of production for twenty years. The information you've requested, pin-outs and such, was included in the published data sheets from that era. Since the data sheets were published they were not proprietary information."
Does anyone besides Highly Liquid know the pin outs for the speak and spell chips, and details of how to access the rom, because John isn't forthcoming with this info. While he is obviously totally awesome, I feel that this info should be made public.
Continuing to look for clues with Texas Instruments, I asked if I could get in contact with old engineers, and they simply said "the engineers who worked on the Speak & Spell products have either retired or moved on to other divisions of TI so we are not in a position to contact them."
Too bad. I'm continuing to do a few experiments to figure this thing out. If I crack it I'll release the info so people can do their own midi for the speak and spell. If anyone would like to assist in this open source type project, please send me a PM. |
|
|
Back to top
|
|
 |
mosc
Site Admin

Joined: Jan 31, 2003 Posts: 18260 Location: Durham, NC
Audio files: 228
G2 patch files: 60
|
Posted: Fri Jun 09, 2006 10:38 am Post subject:
|
 |
|
I know the guy who developed the chip. I'll contact him, but I can't promiss anything. He's a professor at a major engineering school these days. What is the chip id you want? _________________ --Howard
my music and other stuff |
|
|
Back to top
|
|
 |
Pehr

Joined: Aug 14, 2005 Posts: 1307 Location: Björkvik, Sweden
Audio files: 2
|
Posted: Fri Jun 09, 2006 2:13 pm Post subject:
|
 |
|
| mister wrote: | | highly liquid has been doing this for a while and he is a bad ass. I consider this bending because you are going beyond the traditional purpose of the machine and the sounds it makes are straight out glitch and noise, just like a bent speak! anyone who loves the chaos of circuit bending and the control of MIDI should definitely check out highly liquid. thanks for bringing up this topic PEHR! |
Yes I agree that it isn't true circuit bending. But hey! I know I'd like to control a S&S with my MC-505 Well, that's when I can afford one from eBayof course >_< _________________ YouTube channel
flickr photostream
http://loxodrom.blogspot.com
http://www.garageband.com/artist/loxodrom
http://soundcloud.com/loxodrom |
|
|
Back to top
|
|
 |
Pehr

Joined: Aug 14, 2005 Posts: 1307 Location: Björkvik, Sweden
Audio files: 2
|
Posted: Fri Jun 09, 2006 2:16 pm Post subject:
|
 |
|
| mosc wrote: | | I know the guy who developed the chip. I'll contact him, but I can't promiss anything. He's a professor at a major engineering school these days. What is the chip id you want? |
That's really cool If you can get a pdf of the chips it would be awesome
| Google wrote: | | The Speak & Spell project was started in the year 1976 and created with the TMC 0280 the first one-chip LPC speech synthesizer. Later refinements to the Speak & Spell chips resulted in the TMS 5100, 5200 and 5220 Voice Synthesis Processors for use in commercial products needing synthetic speech voice output from digitally-stored words and phrases. Speech data was stored in in up to sixteen 128K ROM chips (TMC 0350). Plug-in modules increased vocabulary and provided for versions in French, German, Spanish and British English. |
 _________________ YouTube channel
flickr photostream
http://loxodrom.blogspot.com
http://www.garageband.com/artist/loxodrom
http://soundcloud.com/loxodrom |
|
|
Back to top
|
|
 |
jksuperstar

Joined: Aug 20, 2004 Posts: 2503 Location: Denver
Audio files: 1
G2 patch files: 18
|
Posted: Fri Jun 09, 2006 3:22 pm Post subject:
|
 |
|
As far as most of the hacks are concerned (in terms of the highly liquid midi mod), it is simply a seperate pc board, with a microcontroller doing the MIDI input. Then it controls the common (and readily available) hacks found on the 'net via software, instead of most methods that use switches and what not. Are you looking for more accurate hacks, or your own method of MIDI hack?
In essence, I don't think you need more info to do even what highly liquid is doing. |
|
|
Back to top
|
|
 |
mi_dach

Joined: Dec 17, 2005 Posts: 133 Location: Sweden
|
Posted: Sat Jun 10, 2006 2:56 am Post subject:
|
 |
|
The Highly Liquid midi mod accesses the bus to the rom chip, and sends data, which gives the microcontroller the option to play sounds or combination/sequence of sounds. f you map all these to midi notes, you can then have it say new words by sending appropriate midi sequences. Highly Liquid has done some fine engineering in figuring this out!
This article provides a good starting point, but sadly I havn't got the time at the moment for all the experiments required to figure it all out. |
|
|
Back to top
|
|
 |
electrospeaker
Joined: Jun 30, 2006 Posts: 49 Location: Norway
|
Posted: Fri Jun 30, 2006 4:44 am Post subject:
|
 |
|
Wow! Interesting discussion. I've been searching for a forum where I can discuss the Speak & Spell.
I'm currently trying to create a line-out audio output from my Speak & Spell Compact which isn't as easy as it sounds (no pun intended)
It has a built in piezo-speaker, and by simply removing the speaker and connecting it to a speaker I get this awful, almost distorted sound. I've tried adapting the circuit in various ways (resistors in parallel to re-create a load on the circuit now that the speaker is gone, and resistors in series to attenuate the output), but nothing helps.
I got a bit closer by insering a 10nF capacitor in series, which removed a lot of the lower frequency "dirt" in the sound, but also removed other frequencies.
The thing is that the "Compact" uses the same speech chip as the regular Speak & Spell (the Texas instruments TMC-0281), and the latter toy has a much better sound (it also has a headphone jack), so I'm assuming that the rest of the circuitry somehow "shapes" the sound to make it more suitable for the (limited frequency-range) piezo-speaker, unlike the regular Speak & Spell which has a normal speaker and a headphone output.
After a lot of searching and asking around I finally got hold of a (badly scanned) schematic for the Speak & Spell (not the "Compact") as well as some information from another circuit-bender, confirming that pins 11 and 12 of the TMC-0281 delivered the audio output. But here's the strange part: on the "Compact" I cut the circuit traces from those pins so there would be no more connection with the rest of the circuit, then I fed pins 11 and 12 directly into my mixer/amp. I would think that the sound would be a lot better (or at least different), but no... the audio level was a lot lower, but just as bad!
So I thought I'd compare it with my regular Speak & Spell, doing the same thing there (cutting the traces and tapping audio directly from those pins). The sound was exactly the same as before (clean and good), but surprisingly I could still hear audio coming out of the on-board speaker!!?!?!
I assured that the traces were indeed cut, but either there are traces on the other side leading from those pins which I can't easily see (because the IC is blocking them), or the pins 11 and 12 of the speech-synthesis chip (TMC-0281) are only partly responsible for the audio.
Another explanation could be that filtering/shaping of the sound somehow occurs before the output.
I'm puzzled!
I'm going to go through the Speak & Spell schematic again (it's better than nothing, but what I could really need is a schematic for the "Compact"), but does anyone here have any ideas?
Also, I found it very interesting to read what Mosc said about knowing the person who designed the chip for the Speak & Spell. Did he design the Speak & Spell as well? (hey, that rhymed!).
I'd really like to get in touch with him or anyone else related to the Speak & Spell project. I've contacted Texas Instruments, but just as Mi-dach experienced they can't be of any help  |
|
|
Back to top
|
|
 |
mi_dach

Joined: Dec 17, 2005 Posts: 133 Location: Sweden
|
Posted: Fri Jun 30, 2006 7:55 am Post subject:
|
 |
|
As far as I was aware, there was no schematic or pinout information available at all, perhaps you can post the schematic you already have?
The only thing I've found is this old article that briefly describes the bus protocol. |
|
|
Back to top
|
|
 |
electrospeaker
Joined: Jun 30, 2006 Posts: 49 Location: Norway
|
Posted: Fri Jun 30, 2006 3:44 pm Post subject:
|
 |
|
You never know. Someone might just have that information stashed away somewhere.
As for the Speak & Spell schematic.. it's available in the "files" section of the Yahoo group called benders. You have to be a member in order to access those files, so I've taken the liberty to temporarily upload it here for a limited amount of time.
(NOTE: removed, but file has been uploaded to the next message in this thread) -see below.
It looks like it was scanned from a book called Electronically speaking: computer speech generation by John P. Cater.
If anyone has (or gets hold of) that book, a better scan would be very much appreciated!
Who knows, perhaps the book has other circuit diagrams as well! Last edited by electrospeaker on Mon Dec 04, 2006 4:22 am; edited 1 time in total |
|
|
Back to top
|
|
 |
mosc
Site Admin

Joined: Jan 31, 2003 Posts: 18260 Location: Durham, NC
Audio files: 228
G2 patch files: 60
|
|
|
Back to top
|
|
 |
electri-fire
Joined: Jul 26, 2006 Posts: 536 Location: Dordrecht NL
Audio files: 4
G2 patch files: 4
|
Posted: Wed Jul 26, 2006 4:18 pm Post subject:
"Old well known" speak schematic resolution |
 |
|
It's late, i'm drunk....
I do however have the impression the presented speak schematic is of better resolution than the one presently available in various locations.
PLEASE correct me if I'm wrong. If not I'll eventually take up the chore of updating this file at benders/benders anonymous/ electri-fire.com and any others I can find.
(hope I'll have some braincells left though.... ) |
|
|
Back to top
|
|
 |
ian-s

Joined: Apr 01, 2004 Posts: 2672 Location: Auckland, New Zealand
Audio files: 42
G2 patch files: 626
|
Posted: Wed Jul 26, 2006 5:02 pm Post subject:
|
 |
|
| Has anyone heard a software speech sythesiser that sounds close to Stephen Hawking's unit? It is a very old hardware unit made by speech plus. |
|
|
Back to top
|
|
 |
stevenclements

Joined: Nov 16, 2005 Posts: 74 Location: Edmonton, Alberta
|
Posted: Sun Jul 30, 2006 10:20 am Post subject:
|
 |
|
I just ordered the Reed Gaza book
Circuit-Bending: Build Your Own Alien Instruments
Reed Ghazala
ISBN: 0-7645-8887-7
Paperback
450 pages
August 2005
I hope this gives me some guidance... I'm suprized with all th e bent S&S unit out there and all the people into this that there's no much sharing of information on tweaks and bends... _________________ http://soundcloud.com/pilfermusic
http://www.gearjunkies-userstudios.com/studio.php?stid=221
life is a treasure... dig it! |
|
|
Back to top
|
|
 |
Pehr

Joined: Aug 14, 2005 Posts: 1307 Location: Björkvik, Sweden
Audio files: 2
|
|
|
Back to top
|
|
 |
stevenclements

Joined: Nov 16, 2005 Posts: 74 Location: Edmonton, Alberta
|
|
|
Back to top
|
|
 |
bigtex

Joined: Mar 30, 2006 Posts: 323 Location: Cupertino, California
|
Posted: Wed Aug 16, 2006 5:33 pm Post subject:
|
 |
|
| g2ian wrote: | | Has anyone heard a software speech sythesiser that sounds close to Stephen Hawking's unit? It is a very old hardware unit made by speech plus. |
The original Macintalk sounded just like it. I think the voice is still available as "Fred" even to this day. Or at least the last time I used speech synthesis on a Mac (MacOS 8 days, I think). There was even a program called VocalWriter designed to make the Macintalk voices sing. It was a full MIDI sequencer, too. You could do barbershop quartet style stuff with multi-part harmonies. Quite fun. |
|
|
Back to top
|
|
 |
dnny

Joined: Mar 12, 2005 Posts: 519 Location: Helsinki, Finland
Audio files: 8
|
|
|
Back to top
|
|
 |
djscoto
Joined: Mar 07, 2007 Posts: 1 Location: Chicago
|
|
|
Back to top
|
|
 |
Mikmo
Joined: Dec 01, 2005 Posts: 150 Location: Copenhagen - Denmark
|
Posted: Sun Mar 11, 2007 3:53 am Post subject:
|
 |
|
on some of the very old s&s models, the ones with real buttons not membrane keypad, you need a small transformer to make a line out.
As far as i remember it's described in the Ghazala book _________________ Stay Cool
Mikael
http://www.mikmo.dk |
|
|
Back to top
|
|
 |
xray303
Joined: Feb 28, 2012 Posts: 1 Location: gva
|
Posted: Tue Feb 28, 2012 7:54 am Post subject:
|
 |
|
Hi,
I just released the code for a little project : Arduino Midi Speak and Spell.
It permit to control via Midi the speak and spell like the Highly liquid kit.
As it is open source, code and schematics provided, you can modify the code as you want and add feature like midi controlled pitch for example or anything.
If you are interested, you can check this at : http://noizeinabox.blogspot.com/
That's it. |
|
|
Back to top
|
|
 |
|