Sorry your browser is not supported!

I just rewatched the 1970 film Colossus: The Forbin Project and have been itching to recreate the voice from that film. You can listen to it here: http://www.youtube.com/watch?v=EMpeiYgLRdo

After doing some research I haven't found any way to accurately create that voice using existing TTS systems. I've gotten poor results modifying the settings of pre-existing voices and tend to end up with nicer/kinder sounding voices no matter what I do.

One individual I talked to suggested that I should try to emulate the old VS100 chip which is the core chip in the TRS-80 Model 1 Voice Synthesizer that was allegedly used to create the voice for the movie. Unfortunately, I can't seem to find a hardware diagram of the chip's internals. I do know that it used some analog tech that might be hard to properly emulate in software.

So, has anyone here ever coded a voice synthesizer from scratch? A friend of mine in college wrote one (took him 5-6 months and it sounded awful). Unfortunately, I have no method of contacting him anymore. I've been looking for algorithms online and so far have found very little that doesn't involve "Download this pre-existing tool and use it."

This is a learning experience for me. In this case I WANT to reinvent the wheel, so to speak. The problem is, I'm not having any luck finding adequate research material. Any help would be appreciated. However, since this is more for "edutainment" it isn't necessarily important if nobody can help.

Cheers,
Frank

Well, I know that FL Studio contains a TTS voice that sounds very similar to that (I used it ages ago), You could try downloading that.

Thanks blobby, I'll check that out. I don't know why I have this urge lately to program a voice synthesizer from scratch, but I'll start with FL Studio. Thanks for the heads up.

Cheers,
Frank