config.json

{
  "text-to-speech_interface": "TransformerTTS",
  "vocoder_interface": "HiFIGAN",
  "vocoder_model_id": "speechbrain/tts-hifigan-ljspeech"
}