Using audio-only files in a Lora dataset

r/StableDiffusion
AI Research

Is there a way to (also) use audio-only files to train a person's voice on a LTX character Lora on AI-Toolkit or some other