How close are we to real-time generated speech?

Is the game industry any closer to achieving in-game speech in its purest interactive form? How close are we to the ‘Holy Grail’ of typing words and hearing back believable voices?

This month’s issue of Develop approaches that very question, with an interview with Paul Taylor, the CEO of Cambridge-based Phonetic Arts.

The interview – which you can read here – reveals how its newest tool PA Studio is taking the first tentative steps toward real-time speech.

“Essentially what PA Studio does is that it builds a statistical model of the voice – that’s where the magic is,” Taylor explains.

Taylor states that the tech isn’t yet advanced enough to make that technological leap into believable real-time audio. However, However, the tech can provide other interesting opportunities for audio teams.

Click here to read more.