Recently I have been dabbling with speech synthesis in Webkit after I found a really cool CodePen by Matt West where I could write in some text and get Chrome to speak it out loud for me. I initially thought this would be an awesome little tool for some kind of assistant or accessibility tool but later I use the code to adapt it to a check-in system in my current workplace that allows a member of staff to tap-in and out with a key fob on a Raspberry Pi and it would give verbal feedback. It took me around 30 minutes to dissect the code that I needed, I couldn't find a library or plugin online that did exactly what I needed and thought that building something to facilitate speech in the browser would an interesting little project.
And then there was light. After an evening of playing with speech synthesis in Chrome and Safari I managed to get a working plugin together called Speak.js - the idea being that you can quick and easily get your browser speaking with very little effort but with the capability to change voices, pitch, rate and volume; have a look at the Github repository and let me know your thoughts via Twitter.