Voice Failure?

I translated a short video to test this feature, and although adobe promises "Keep the same voice, tone, cadence, and acoustics as the original video" the translated voice was nothing like the original. 

I also expected that I'd get lip synch. The faq says the AI is able to "generate speaker voices, translate between languages, and adjust new lip movements. " It did one of these things for my video. 

How do I access options? All I see is choose language and generate. 

How do I tweak it the output? Do I just start from scratch and translate again for another 850 credits?