Stephen Hawking Voice Generator

Feel free to use the generated audio for any of your projects (commercial or personal). It’s free! Hope it’s useful for you 🙂

This is a simple online tool to convert text into an audio clip of Stephen Hawking’s voice. It’s based on the eSpeak library which was developed by Jonathan Duddington in 1995, and which has since gone through many updates. Hawking’s voice was based of a similar sounding synthesiser from the 1980s, but has also gone through upgrades throughout the years.

This Stephen Hawking voice generator uses the JavaScript port of the eSpeak library which was done by @kripken – many thanks to him! It allows you to generate the robotic audio and then play it and download it as a WAV file.

eSpeak uses "formant synthesis" rather than piecing together pre-recorded clips. This generally allows for a much more fine-grained control over the produced audio, but older versions of formant speech synthesis such as this tend to have a very robotic, monotonic feel. Conversely, the systems that use pre-recorded clips tend to not be monotonic, but instead have unusual and incorrect inflections on words, or whole sentences.

Since he lost his ability to speak in the late 1970s to early 1980s, Stephen Hawking’s text to speech’s is easily recognisable as one of these formant synthesis tts systems.

This system produces audio clips in the WAV format, but if there is enough demand I may be able to produce an MP3 download link too.

It’s probably best to try inputting a small amount of text first, and then increasing it to see how much your computer can handle. The computation time may be quite big for large bodies of text.

I used to have a lot of fun creating funny text to speech phases using Microsoft Sam, so I hope this brings that same experience to others via the medium of this little online tool. If you come across any fun phrases, paste them in the comments section below! I’d also love to hear what you’re using this for. Thanks for using my app! 🙂