Text-to-speech; synthesized podcasts


People often ask me what Microsoft’s doing with text-to-speech voice development.  If you’re interested in this, you may want to look at the “Mulan” project from the speech team at Microsoft Research Asia.

I’ve been listening to various podcasts for months now.  They’re a great idea, and once I found some good quality ‘casts I was interested in they became a wonderful way for me to learn stuff while my eyes and hands are busy doing other things (driving, working around the house, etc).  (Incidentally, I use Doppler, because I find it does a good job of converting MP3s to the .m4b format for my iPod).

So, on the topics of podcasting and TTS, it’s nice to see people experimenting with using speech synthesis to convert RSS text feeds into MP3 files.  I’ve played with Feedspeaker and I’ll probably try Autocast sometime soon.  I can’t help feeling there’s a lot of potential here.


Comments (1)

  1. Greg Smith says:

    If you have a Text-to-Speech engine on your Pocket PC, you may be able to hook it up to FeederReader (an RSS reader and podcatcher for the Pocket PC) to have FeederReader export your message and play the file using a TTS program (of your choice, as long as it can be executed to read a file).

    FeederReader has custom Export commands that you can define to call your TTS engine.

    If you need help, let me know on the FeederReader message board (link is on the help page of http://www.FeederReader.com)

    Greg Smith

    Author, FeederReader – The Pocket PC RSS, podcatcher, videocatcher

    http://www.FeederReader.com – Download on the Road