Pynacea

Pynacea is a voice recognition framework and grammar specification that gives users control over their desktop environment through voice input.

This project is similar to the Natlink/Dragonfly/Aenea stack. However, Pynacea does not hook into Dragon NaturallySpeaking. In exchange for losing the ability to send commands to NaturallySpeaking, this project hopefully gains stability and flexibility in that it does not require access to APIs from any proprietary voice recognition software, and can in fact run alongside any voice recognition program.

Pynacea can communicate with a speech-to-text engine through the pynguest script contained in this repository. It also can run off of streams of text via sockets, a subprocess' stdout, or HTTP requests.

Read the documentation here.

To Do

• Native integration with Windows Speech Recognition
• Support for Mac OS X host machines