> Accent, dialect, and low-resource language adaptation — adapt a base Gemma model to underrepresented voices and languages with your own labeled audio.

Is this for TTS? Have been looking for something to do a local fine tune to get a specific accent