Usage

1. Clone the repo

2. Add your models to the Models directory. Models must use the F5-TTS achitecture.

3. Add reference audio to Data/Audio as well as the transcription of the audio to Data/Transcriptions.

4. From terminal open the project directory and start app with

uv run echobox

5. From there it is pretty simple to use, just select the model configuration... model, reference files etc.. from top menu. Paste your text and click the inference button in the bottom left. When its finished processing the generated .wav file will be loaded into the built in media player.