Welcome to my implementation of a speech-to-text tool for the VRChat textbox.

Disclaimer: You will need a Microsoft Azure Speech-Recognition service. You can find infos about it on Microsoft Azure.
This is a project i started for myself, if you wanna use it or play around with it, you are welcome to do so, but it WILL have bugs and i don't know if i'll even continue develpoment here.
If you want a program that just works out of the box, check out https://vrcstt.com/.

• Look for the newest release on the Release Page and download the .zip
• Unpack it somewhere
• Open the file 'VRCSTT.dll.config'
• Place the SubscriptionKey and the Region of you Azure speech-recognition service inside the 'Value'-fields.
• Run 'VRCSTT.exe'
• Set the slider to the time that should be waited before sending the remaining part of a message, if it is too long to be sent at once. (VRChat can only display 144 characters at once)
• Make sure that OSC is enabled ingame (Circle Menu -> Options -> OSC -> Enabled)

• Follow the normal setup above
• Add an ExpressionParameter to your avatar called "StartVoiceRecognition" (boolean, default false)

Voice recognition will now start any time you set the "StartVoiceRecognition"-Parameter to true. You can do that easily by adding a "Button" to your ExpressionMenu that activates this parameter.

If you have any problems, you can open an issue; don't expect immediate response tho.