VoxCommando offers its users a choice of two different speech recognition engines. The standard Microsoft Desktop Engine is a great choice for those who natively speak one of the 8 languages it supports and who want to have the flexibility of free dictation and voice training profiles. For our global community, we also support Windows Speech Platform 11 (SP), which provides speech recognition in more than 20 different world languages and accents, does not require training, and works better with Bluetooth microphones.
VoxCommando has multiple options for handling when and how it listens for commands, depending on user hardware and preferences.
There are three listening states: on, standby, and off. In Prefix Mode, VC remains in standby and will only try to execute commands if it first hears your special prefix (e.g. “Jarvis”). Users can choose their own prefixes. Prefix Mode is an important way to minimize false positives for those who want to use an open-air microphone.
It’s also possible to use events to switch between listening states and to turn system volume down temporarily in order to minimize background noise. Among other options, events can be generated by a keyboard shortcut, a USB button, a motion sensor, even a z-wave or X10 switch. It’s your choice. Find the best solution for your environment.
More than 150 core actions, and over 30 different plugins allow you to create custom solutions for virtually any situation. Context sensitive groups of commands, custom Python scripting, and an api that supports command line, TCP, UDP, and HTTP control mean there is almost always a way to make VoxCommando control or respond to input from other programs or devices as long as they provide some means of communication.