1 min voice data can also be used to train a good TTS model! (few shot voice cloning)
Publish a package to this repository to see it listed here.