❓ Questions / Help / Support

## ❓ Questions and Help

Is it possible to train your own speaker (vocoder) based on given dataset of audio-s and transcriptions to use it within the pipeline of `silero_model.apply_tts(..., speaker=custom_speaker)`? 
