Text to Speech VST : VST Speek !

Techno is reactivated

Techno is reactivated

VST Speek is a free vocal synthesis VST plugin for recreating the old skool robotic text to speech we all love. Choose a preset, type any text and let’s speak !

download VST Speek 1.3 for Windows 32 bit

download VST Speek 1.3 for windows 64 bit

VST Speek + reverb


– fix for crash in Cubase when selecting preset from drop down menu
– add hack for Presonus Studio One : text is now editable
– fix memory leaks
– MAC OS X version !!! => see https://blog.wavosaur.com/text-to-speech-vst-now-for-mac-vst-speek-au-speek-audiounit/

– now the  hack for getting keyboard events in Ableton is less dirty 🙂
– fixed potential bug on close plugin

– 64 bit version for Windows
– Fixed bug with audio buffer
– Added hack for Ableton Live 9 : plugin should get keyboard keys input now
– Added support for VST Key (not tested ! we don’t have Cubase)
VST Speek text to speech is 100% freeware, the plugin is based on the Software Automatic Mouth (SAM) vocal synthesis software created by Softvoice Inc for the Commodore 64.
The original C64 software has been reverse engineered and ported to C by Sebastian Macke : http://simulationcorner.net/index.php?page=sam

I have adapted then to VST plugin :

  • you can trigger the speech with MIDI keyboard (velocity sensitive)
  • you can use MIDI pitchbend
  • you can change original pitch or mouth/throat parameters while it’s speaking

This is not a vocal synth like “Vocaloïd”, it’s intended to do crunchy robotic lofi techno voices!
It’s not intended for good quality speech synthesis too, better speech synthesis softwares exist now.

Please comment and vote for the most wanted feature / development for this VST plugin !

– 64 bit version
– MAC version
– ability to assign different text to MIDI notes
– ability to read longer text (and / or read from text file)
– assign pitchbend or midi CC to voice parameters.
– Android version ?
– more parameters

Don’t hesitate to report any bug you found.



Text to speech vocal synthesis has been used in some famous (and less famous!)  rave/techno songs from the 90’s. The robot voices were popular in the nineties.
The funkyest speech synthesis sotfwares at this time were :

The Amiga 500 say program :
it used narrator & translator library to produce nice computerized robot voices ! If you are not a happy Amiga 500 owner you can use an Amiga emulator : WinUAE

The Commodore 64 Software Automatic Mouth (SAM) :
you can use the brand new VST Speek or the S.A.M simulation program (command line executable for Windows) from Sebastian Macke  : SAM for window

The Atari ST Speech  :
that’s the one in U96 – Das Boot ! you can use the Steem Atari emulator + stspeech.tos
Here’s a port for Windows, command line application : stspeech.exe

Apple MacinTalk

They were all developped by Softvoice inc.

We must not forget the Speak n Spell toy by Texas Instrument !
This was a great piece of talking machine, that is still circuit-bended to death nowadays.


Here’s some 90’s music tracks that used vocal synthesis (Amiga or Atari) :

U96 – Das Boot :

U96 – Das Boot / Kennedy (megamix):

Snap – The Ex-Terminator :

Quadrophonia – Quadophonia :

Signal Aout 42 – I Want To Push :

Marcel Hol – Serious Beats Vol.4 Mega Dance Party Mix (1992) :

The Deadkirks – Mr. Kirk, your Son is Dead :

General Base – Apache & Base Of Love :

The Scientist – Spiral Symphony

Rave Crusaders – Energy Overload

Pleasure Game – Activez Les Plaisirs (Amnesia Robot Mix) :
http://www.bide-et-musique.com/song/11963.html  Daktylus is reactivated !!!

+ update !

Romanthony “The Wanderer” (Journey Man Thump)

Smart Systems – The Tingler

1 2 3 techno ! maximum velocity ! reactivate ! maximum overload energiiiiiiize all system reactivate move !

116 thoughts on “Text to Speech VST : VST Speek !

    1. aum hren

      thx for sharing
      i dl windows version, from above link:
      VST Speek 1.3 for windows 64 bit
      unzipped etc
      now, documentation:
      VST Speek documentation.pdf
      has a link to:
      The complete phoneme alphabet and explanations can be found here :
      this link does not work for me
      i have found the phoneme alphabet at:
      1) http://www.apple-iigs.info/newdoc/sam.pdf
      2) https://github.com/s-macke/SAM/wiki/Phonetic-Alphabet

      if it makes sense to you, please fix the link, or, better, include (copy-paste, but change W whale -> WH whale) s-macke’s sam wiki text in ‘vst speek documentation.pdf’ file

      have fun

      s love nia

  1. Simon

    I’d love to have pitchbend. Then the ability to assign different text to MIDI notes. More parameters are always a good thing, too. 🙂

    1. wavosaur

      I had this problem in “minihost” : it was because there was the “use computer keyboard as midi keyboard” option activated : when unchecking this option i had the VST speek working again.
      Maybe this is the same kind of problem with live 9.

      1. along

        i have the same problem on ableton live 8 for windows 7. i deactivated the “computer MIDI keyboard” but i can’t type the text in the box.
        the sound is wonderful!

  2. Phil

    Nice one !
    It is really funny to have this kind of speech synth in a plugin format….much more than running emulators 😉

    Thanks !

  3. No Mac

    I’m loving it. …..
    Thank you for this.

    If you’re a MAC user, you should already know that disappointment is a normal thing.

  4. Jim

    I came up with a work round for Live. Use blocknote to write your text, copy from blocknote then paste into the text field. I recorded the result then clipped the noise from before and after the speech.

  5. wavosaur Post author


    I have fixed a bug with audio buffers, it occured in Ableton Live 9, Nuendo 4 and surely many other hosts!
    I did a dirty hack to get the textfield responsive in Ableton : it’s a dirty job, but someone got to do it

    64 bit version has been tested in Live 9, VSTHost by Hermann Seib and Wavosaur !

    I also implemented some VST Key feature, but this is not tested, don’t hesitate to comment if the keyboard bugs are fixed with Nuendo & Cubase with this version.

    keep the good vibe!

  6. anon

    I love that sound but they crashes on closing, dialog says R6025 – pure virtual function call.
    Both 32bit and 64bit v1.2, tested on savihost. you may forgot writting ‘virtual’ for some
    destructor(s) or something?
    Sorry for my poor english.

      1. anon

        OMG pen_tablet.dll for WACOM tablet is doing wrong. Looks its not your fault.
        This is call stack captured using ollydbg,

        0018F99C 75D6FD57 USER32.MessageBoxExW USER32.75D6FD52 0018F998
        0018F9A0 00000000 hOwner = NULL
        0018F9A4 68F10998 Text = “Runtime Error!..Program: savihost.exe..R6025..- pure virtual function call..”
        0018F9A8 68EF33E8 Title = “Microsoft Visual C++ Runtime Library”
        0018F9AC 00012010 Style = MB_OK|MB_ICONHAND|MB_TASKMODAL|10000
        0018F9B0 00000000 LanguageID = 0 (LANG_NEUTRAL)
        0018F9B8 68EA4605 Includes USER32.75D6FD57 Pen_Tabl.68EA4603 0018F9B4

  7. Actvi8

    very nice plugin ! Only in cubase 5 (64 bit), I can not type a space between the words. Also it crashes when selecting a preset. It would be cool that the parameters can be changed with a midi controller so you can record them during a song.

    1. Wavosaur Post author

      Hi !
      We’re about to update to v1.3, for typing text in Cubase i’ll apply the same hack we used for Ableton.
      I have not find a host able to reproduce the crash on preset change :
      does it happen everytime ?
      does it happen when GUI is open ?
      does it happen when GUI is closed ?
      does it happen while playing and or when stopped ?

      parameters should be automatable : pitch speed mouth & throat can be automated (works in other hosts)

      1. Actvi8

        hello great to here an updates is on his way 🙂
        The crash only happens when you click on the dropdown of presets in the vst plugin menu. when you load a preset it opens the windows eplorer dialog and then you can select a preset and the preset loads without a crash. So only the dropdown selection causes a crash of the entire DAW (cubase 5).

        1. Wavosaur Post author

          finally i installed the Cubase Element 7 trial : 30 days should be sufficient for finding the bug!

          i was able to trigger some ugly problems when using the dropdown selection.
          it seems it’s a thread issue, a fix should come on monday 🙂 !

        1. Wavosaur Post author

          What host are you using ?
          I made some fixes for host that steal the keyboard events (Ableton, Presonus Studio 1)…
          some others host have options to send the keyboard event to the plugin (Renoise if i remember correctly, and maybe Reaper)

  8. Aram

    Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac Mac


    But big thanks for letting us rock with this on Windows too of course !

      1. Wavosaur Post author

        Hi Frankie
        can you please send the crash report to contact@wavosaur.com ?
        We were able to debug the VST version in Cubase with the help of crash report

        also, what version of the plugin did you use ? (AU or VST, 32 or 64 bit) and what is your OS ? Snow Leopard or Lion or Mountain Lion or Mavericks

        thank you for your bug report !

    1. Wavosaur Post author

      Nice track !
      i don’ think i’ll add other voices to this plugin.
      i have another project with more natural voice, but based on the microsoft tts engine, it was the first attempt to get this robotic text to speech (ms tts + bt crush) but it wasn’t satisfying

      maybe i’ll release this other project (but windows only!) if some people feel the need for more natural voice

  9. 74smoky

    Nice tool! It would be very cool if the plugin would support a mapping mode.

    In this mode it would be able to map a word or a sentence to a distinct key with a distinct tone/pitch, e.g.

    {D4;F3}Liberate this now

    1) Note D4 and all above are the spoken sentence “Liberate this now” in tone F3 and all tones above respectively
    2) Notes C4 and C#4 are the spoken word “Exterminate” in tone C3 and C#3 respectively
    3) Notes from C3 to B3 are the spoken word “Aggregate” in tone C3 to B3 respectively

    So the syntax states:
    {Mapping starting point;Tone to starting point}Spoken word

    1. Wavosaur Post author


      quel OS utilisez vous ? Windows XP, Windows Vista, Windows 7, Windows 8, Mac OS X (quelle version?), Linux ? OS9 ? AmigaOS ? autre ?

      quel DAW / sequencer / host utilisez-vous ? Ableton ?, Cubase ?, Sonar ?, Mulab ? Traktion ? EenrgyXT ? Minihost ? Chainer ? Studio one ? Muzys ? FL Studio ? Orion ? Cantabile ? Nuendo ? autre ?

      le host utilisé est une application 32 bits ou 64 bits ?

      En général il suffit de copier la dll dans votre dossier VST plugins et le plugin apparait au prochain rescan du host (en général il le fait au demarrage de l’application)

      1. djjmi70

        ha super ! il y a quelqu un heu je suis sous windows 7 j utilse acid pro 7 et adobe audition 3.0

        mon windows est en 32 bits

        en tout cas je suis fort interesse car au vu de la demo avec cette voix type u96 , elle est des tres bonne qualite!
        en vous remerciant de votre aide

  10. Philbert

    I am unsure if anyone even still comes to this page, though I am hoping I am wrong since I have an issue with this plugin, when I load it through wavosaur and view it, it shows up and I see the whole UI (or what ever it would be called) that is shown in the image, though when I hit the speak button after typing in words I receive no sound from it at all. Does anyone have any clue why this is the case?

    1. Wavosaur Post author

      Wavosaur is more an audio editor than a sequencer.
      but if you want, you can use VSTSpeek inside Wavosaur : first create an empty file (stereo, 44100), check the “processing” checkbox, be sure to be in “loop playing mode” and then hit “play”.
      Wavosaur has to be playing for processing plugins.

      you can also do that :

      create an empty file
      load VSTSpeek
      type your text, press the “sPEAK!” button then hit “apply” in Wavosaur

  11. Logic-Guru

    This Plugin does NOT produce any Sound in Logic 5.51 Win!
    Insert and GUI are all OK, but NOT AUDIO OUTPUT SIGNAL!!!

    Test In SaviHost and “Seib VST-Host” works fine, and works also in SONAR8 at an Coop-Studio…

    Sorry, i’m a “LOGIC ONLY” USER in my “PC Project Studio”!!!
    In my Main “MAC Studio” are using Logic and DP with TDM Plugins ONLY 😉

    greetz: Frank from NRW/Germany – sorry for my poor english…

  12. Paul

    Fantastic application. Thank you for letting me use it for free, though you could have made a bit of cash putting it at $5 or something like that. I was about to go through the hassle of setting up an emulator lol.

  13. The cat

    exe files are better then dll files you dumbass you going to fucking reply to this fucking comment or i fucking start a copony and fucking fuck over copyright and fucking bitch slap all of you fags!!!!!!!!!!!!!!!!!!!!!!!!

  14. Jeroen

    Just installed the VST 32 version in Image Line Minihost Modular (32bit) but I can’t seem to input text with my laptop keys. I have Windows 7 Home Premium SP1 32bit. Is there a workaround for this issue?

  15. Robin

    Love it. Small problem, I cant edit the text. I’m using MPC 2.5.0 with VIP 3.1.1 on Windows 10. I would love to be able to use this effectively. Best wishes.

  16. MannyG

    How much text can the vst hold before pausing. I’m using live 9 and it only reads like two sentences and then stops. I’m just wandering if there’s a cap on how many words or characters it can read because if there is. it shouldn’t allow me to add so much text.

  17. AliceD

    ziemlich cooles Vst, aber wie kann ich es in Cubase aufnehmen?
    zum anderen, hat wer nen Tip, wie ich ihn deutsch sprechen lassen kann?

    1. AliceD

      realy nice vst, but i got 2 questions, how can i record it in Cubase, it doesn´t react to midi notes and is therte a way tolet him speek in german?

  18. Me

    Hi, using VSTSpeek 64bit on Windows 10 and Reaper 6.28 64bit.

    The only key that doesn’t work on my keyboard is the SPACE bar, I can type things into the textbox just fine, except the space bar doesn’t work, also there is no “Dirty Hack” button

  19. alexander


    unfortunately there are some graphic errors (at least in studio one). maybe in other daw’s too. there will be a new version. I would also buy the plugin.


  20. Stefan "ensonic" Sauer

    Sadly the linux vst does not work in bitwig studio (anymore):
    “Plugin did not create it’s window inside the plugin window”
    it would be awesome to have this one on github, so that I could contribute a fix.


Leave a Reply

Your email address will not be published. Required fields are marked *