TTS_APP

Installing

• Install Python

  • https://www.python.org/
  • Click Download and select your Operating System (NOTE: Latest version should work, but was tested with 3.13 and 3.12)
  • Install Python (NOTE: For Windows you can use https://www.pythonguis.com/installation/install-tkinter-windows/ as a Guide)

• (Windows) Install the Microsoft Visual C++ package

  • Download the Lastest supported version from https://learn.microsoft.com/en-us/cpp/windows/latest-supported-vc-redist?view=msvc-170
  • Install the package

• Install Application Dependicies

  • Open a Terminal and navigate to the Application directory
  • Run "pip install -r requirements.txt"
  • NOTE: You may see errors on missing dependencies (aka compiler) and on Windows a link is provided on where to download it
  • Download a Voice into the Voices Folder
    • Samples and Voices can be found at https://rhasspy.github.io/piper-samples/
    • Click the "Download" Link on the chosen voice
    • Download both the .onnx and .json files to the Voices directory in the application

Running the Application

• Running the Application
  • Open a Terminal and navigate to the Application directory
  • run "python main.py"
  • A GUI should pop up

Usage

• Usage
  • Choose an output device for the TTS to play through
  • Choose a voice for the TTS
  • Enter the text you wish to play
  • Click "Play TTS"

Setting up a Live2D Model to mouth the TTS

• Having VTube model "mouth" the TTS
  • Recommended: Install VB-CABLE Virtual Audio Device
    • https://vb-audio.com/Cable/
    • This creates a fake Audio device which the application can interface with and the output can be added to VTube Studio and OBS
  • Open VTube Studio
    • Recommende: I recommend making a copy of the model before making changes. This way your have your normal mode and a TTS model
    • Click "Settings"
    • Scroll down to "Microphone Settings" and click "Use microphone
    • Select "Microphone" and choose either your Microphone or "CABLE In 16 Ch (VB-Audio Virtua" if you installed VB-CABLE)"
    • Click "Model Settings"
    • Find the "Mouth Smile" setting (This may not be the exact name as the rigger can choose any name)
    • Change the Input to "VoiceFrequency"
    • Find the "Mouth Open" setting (This may not be the exact name as the rigger can choose any name)
    • Change the Input to "VoiceVolume"