February 25, 2009. Text-to-speech (TTS) programs have been in use for a number of years, and they’re improving. As Roy Blount says in an op-ed in today’s New York Times, Kindle 2′s TTS isn’t Jim Dale reading “Harry Potter,” but it’s listenable. There’s no need to take our word for it; have a listen to the sample below. To give you an idea of how this technology is improving, we’ve uploaded a couple samples of TTS available from Apple’s Macintosh operating system released in 2005 — the default voice (“Fred”) and the best-sounding one, to our ears (“Vicki”). (We chose Apple as a reasonable benchmark of where the technology stood four years ago because its TTS software is generally considered quite good.)
IBM is one of the companies aggressively involved in developing this technology. For an example of its work, go to the IBM Research site and try its “expressive” samples.