Introducing ShoeVox - Voice Control for Your PC Multimedia
I bought a Kinect back in November and I love it. I think games like Dance Central and Kinect Sports are great, and that it absolutely will continue to expand the world of gaming. However, as much as I use it for gaming, I think it has even more potential with its voice control capabilities. Being able to navigate through trailers or pause my movie without picking up a controller is fantastic. Since I watch a lot of media on my PC, I started thinking with my friend Max how we could bring the experience to Windows. Today, we’re ready to share ShoeVox!
ShoeVox brings the ability to control your multimedia with your voice to your Windows PC. Do you ever lean back in your chair and watch a TV show? Do you ever play music while doing something else in your room? Do you have a Home Theater PC in your living room? With ShoeVox, you don’t need to reach for a mouse and keyboard - you can say “PC Pause” and your media will pause. “PC Play” starts it up again. Other commands like Mute and Next work as well, and the grammar is easily extensible.
ShoeVox is built on the Speech Recognition Engine technology built into the .NET Framework. When it detects a multimedia program that we’ve specified, it starts listening for voice commands - this ensures that system resources are only being used if you’re actually viewing multimedia. When it recognizes a command, ShoeBox controls the running application for you by simulating keyboard input. We currently support Windows Media Center, Windows Media Player, Media Player Classic, VLC, Zune, and iTunes, but this is easily extensible.
ShoeVox is named after the IBM Shoebox, a device demonstrated by IBM in the early 1960s that was one of the first to do speech recognition. Though speech recognition has progressed greatly since then, it’s still a hard problem. We look forward to integrating the forthcoming Kinect SDK if it includes voice recognition APIs. We’ve considered implementing support for Flash (Hulu, YouTube) and Silverlight (Netflix), potentially through UI automation. We would also like to support additional media programs, such as XBMC or Plex. If you have any interest or expertise in these areas, please consider contacting us or submitting a patch!
Please download ShoeVox and try it out, feedback is appreciated!
Fine Print: ShoeVox is released under the GPL v2 license. Though Max and I both currently work at Microsoft, this program was developed on our own time and resources and does not represent Microsoft in any way.