Anfonwch hwn fel neges destun: Modeling of an Automatic Vision Mixer With Human Characteristics for Multi-Camera Theater Recordings