Currently the only way of creating an AI clone voice is to read that consent statement. I want to train it further though as the clone doesnt sound like me at all. Why can't I upload a further 30 minutes of my voice so that the program has more data points to work with. Surely this is essential?