Voice Control 2.1.1 released and news for version 2.2.0!

This news comes with a little delay: some days ago the version 2.1.1 has been released. It fixes bluetooth handling incompatibilities with some devices, and it include also some little improvements.

Full change log:

• Fixed a bluetooth management incompatibility with some devices.
• Fixed crashes on options opening on devices without bluetooth support.
• Tried to apply a fix for rare crashes on options reopening and rotation on some devices (if someone still getting crashes in such situations, please contact me)   
• Removed the confirmation request when the command is recognized with an accuracy of 100% and the results isn't the first one returned by the Voice Recognition engine.
• Little improvements on date handling and parsing
• Added number announce for incoming calls from callers not in the contact list.

Now some news about the next major release 2.2.0.

This version main goal is to include a simple plugin system for actions, in order to allows other developers to write simple actions like one to turn on and off the wifi, one to adjust up and down the volume and so on.

These days I'm working on the plugin API definition and implementation, writing also a “developer guide” to explain how to write plugins. It is almost done and a draft can be found here: https://www.voicecontrolapp.com/dwiki/doku.php?id=developerguide .

If someone already wants to write a simple plugin, he can join to the beta writing here: https://www.voicecontrolapp.com/forum/viewtopic.php?f=12&t=36 . I will be happy to discuss about APIs and documentations improvements.

Also, I'm making some experiments for an “high requested” feature: the offline voice recognition.

First tests are giving pretty good results, an if everything goes well, I might include it in the next version. However, it will come with a limitation: it will not be usable to handle “free text” dictation, like SMS texts or agenda events name, so the “offline mode” will be usable only for some actions, like the “call” action, the “start application” action and so on.

Now the bad news: initially I will be able to include it only for French, English, Chienese (Mandarin), and maybe Russian languages. The reason is that the offline recognition library needs a file called “Acoustic Model” to perform the voice recognition, and such models aren't available for Italian, German and Spanish languages. I can create them from scratch, but I need a lot of voice records of some texts from a lot of speakers, so i will add a little function inside Voice Control in order to allows voluntairs to read some texts and submit them to me, so I will be able to create these models for missing languages.

Stay tuned!