I've been looking all over the net, and I've found some voice synthesizers, but they either don't work, aren't free, or just have 5 or so voices to choose from... I need to get a synthesizer with which I'm able to edit the voice myself. This is needed for GLaDOS-like robotic character for my game.

I did some searching here on GDSE, and it seems that questions like this often get closed as "not constructive". I believe that this is not the case, as this could help many other people who happen to be searching for the same thing.


closed as too broad by Josh Oct 2 '15 at 14:21

Please edit the question to limit it to a specific problem with enough detail to identify an adequate answer. Avoid asking multiple distinct questions at once. See the How to Ask page for help clarifying this question. If this question can be reworded to fit the rules in the help center, please edit the question.

  • \$\begingroup\$ You know, voice synthesizers are pretty complex things to make. It doesn't suprise me that you can't find any "decent" free ones. \$\endgroup\$ – user9790 Mar 4 '12 at 12:21
  • \$\begingroup\$ Hmm, too bad. :/ \$\endgroup\$ – jcora Mar 4 '12 at 12:40
  • 3
    \$\begingroup\$ If Stephen Hawking has to sound like a robot you can be pretty sure that there isn't a better solution available. \$\endgroup\$ – Patrick Hughes Mar 4 '12 at 19:54
  • \$\begingroup\$ Actually, that's the desired effect! As I said, I'm creating an evil robot AI. \$\endgroup\$ – jcora Mar 4 '12 at 19:55
  • \$\begingroup\$ @PatrickHughes Stephen Hawking's voice could be made to sound better. Hawking doesn't want to sound different, he has stated many times that this voice is so tied to his identity that he doesn't want it changed. If you're looking for a voice that sounds like Hawking, his voice synth is available for free here: github.com/01org/acat/releases (See also: nerdist.com/…) \$\endgroup\$ – Felsir Oct 2 '15 at 6:44

Since you're looking for a non-realistic voice check this out, if Java doesn't frighten you then there is: http://freetts.sourceforge.net/docs/index.php

If it doesn't help then going further back Amiga had a TTS engine built into it that may be available somewhere.

Search terms for further investigation would be "text to speech" or "speech synthesis," anything with "voice" in it tends to go off track.


I know this isn't a direct answer to your question, but I'd like to drop in that GLaDOS's voice isn't synthesized though - it was recorded by an actual voice actress and post-processed. That's certainly another possibility to consider.

And just because I thought it was interesting, here's a video that shows how it's possible to create something that sounds very much like GLaDOS in the games, by taking an unprocessed sound clip recorded by the aforementioned actress, and processing it in a voice manipulation tool such as Melodyne. Most of the changes were removing pitch modulations between words, changing the pitch here and there, and messing with the formant of the sound.

I think the original sound clip could even have been recorded with no pitch variation between words, with pitch changes then being added in the application. Melodyne is not free though but there is a 30 day trial you can try. And there might be a free alternative out there, although I'm unaware.

Another alternative to editing the voice like this, is to use software that applies the changes as a whole to your sound clip, and automatically makes it sound like a robot. I'm thinking for instance about using some sort of Vocoder VST, which you should be able to find by googling a bit.

  • \$\begingroup\$ Well, I think that voice acting is out of question. My accent will probably sound weird, because English is not my mother tounge... \$\endgroup\$ – jcora Mar 5 '12 at 22:17

If you're ok with a singsong quality, UTAU. It's decent for a free synthesizer. Also Sinsy. For UTAU, countless tutorials for use are on the web, however it is Japanese and has an English patch, and MANY voicebanks( librarys of samples of recorded voices for concatantive synthesis) Sinsy is on the web, but requires music XML files, which you can make with free software like Musescore. It's not easy though- hence the free part. CeVio is another one, a singthesizer that also has a speechsizer but it's only in Japanese and currently has no released voices in any other language!

As for Stephen Hawking, he could change his voice at any time for better options, but it is part of his idenity now, and at the time he got it- it was the best available. There is better out there though.


Not the answer you're looking for? Browse other questions tagged or ask your own question.