Check out this new earpiece/microphone Intel is developing. This might be a very cool way of controlling voxcommando. Hopefully it's an open API so someone can develop a plugin for it.
http://www.engadget.com/2014/01/06/jarvis-is-intels-smart-earpiece/
If understand it correct, this earbone is a bluetooth headset - I've tested different bluetooth headsets from 10 to 120$ and no one of them works with speech recognition under windows good. The audio compression is too high.
I thought it was an autonomous "smart" device with a fancy intel cpu onboard...
Anyway, I predict 2 things on this one:
1 - This will be a big fail
2 - It will not be "open" in any way.
I think some of the prototypes Intel unveiled this week sound interesting and promising. The next few years should be pretty fun.
It's true that thus far Bluetooth has not been a good match for VC, but that may change in the future. For example, James has mentioned that when VC2 is eventually released, users will have the option of using Windows Speech Platform 11 instead of the built-in MS speech engine. It's possible that WSP 11 does a better job with Bluetooth compression than the regular speech engine. (I don't know if it does, but it's something to investigate.)
That said, whether small, independent developers like James/VoxCommando will be offered a reasonable means to interface with these higher end "smart" technologies isn't as clear. Just look at how cost-prohibitive Apple makes it to develop for iOs.
I can confirm that I am working on VC 2.0 and that it will allow users to choose between System.Speech engine (which is what VC currently uses) and the Speech Platform SDK 11, which users can install.
Kalle has done some tests and it does look promising for bluetooth devices, but more testing will need to be done to say for sure if bluetooth will be good enough for most users.
The main advantage of the Speech Platform is that it will support many more languages, and includes some free TTS voices for many languages as well. These are not the best sounding voices necessarily, but they are free, and at least as good sounding as Anna I think.
Speech platform should also work with no training. In fact it does not allow training at all. It also does not allow for dictation, so you won't be able to use it for web searches etc.
I will be charging users a fee to upgrade from VC to VC 2.0, except for the early adopters that bought lifetime licenses, back in the early days ;)
Yes first tests confirm that the speechengine platform sdk11 works better with bluetooth as the system speech engine. I think one of the biggest advantage is - we do not need a Ultimate or Pro version of windows to get different language packages. :D