Preparation

If your system is not using Pipewire, the software QjackCtl must be installed. Execute it and configure the sample rate to 48000 Hz:

  1. Click on "Setup...".
  2. Set Settings / Parameters / Sample Rate: 48000.
  3. Click on "Save".
  4. Close the "Setup" window.
  5. Click on "Stop".
  6. Click on "Start".

The sample rate is defined by the parameter "output_rate" in the file vtm.txt, located in data/voice/english/*/vtm.txt.

Copy the data files contained in the directory data/ (from the source package of gama_tts) to another directory. The user must have write access to this directory, which will be used in the quickstart.

Run the program gama_tts_editor.

Quickstart

  1. In the main window, click on File / Open and select the directory data/voice/english/5_male.
  2. In the main window, click on "Synthesis window".
  3. In the synthesis window, enter an english text in the field "Text". Click on "Parse". The text will be synthesized.
  4. In the synthesis window, set the "Tempo" field to 0.5. Click on "Synthesize". The text will be synthesized in a slower pace.
  5. In the synthesis window, set the "Tempo" field to 2.0. Click on "Synthesize". The text will be synthesized in a faster pace.
  6. In the main window, click on "Interactive VTM" (interactive vocal tract model). Click on "(Re)start". In the "Dynamic Parameters" area, change the slider for "Glot Pitch". In the slider for "R8" (mouth radius), move the value from zero to the center value and back to zero, repeatedly and as fast as possible. You should hear the sound "ba". Now set the slider for "Velum" to 0.25. Move again the slider for "R8". You should hear the sound "ma". Click on "Stop".