In this part, we’re highlighting what the text to speech AI can do, particularly in expressing a variety of emotions. Keep in mind the following to get similar results:
Clone voices tend to have a better emotional range than pre-made or voice-designed ones. So, we often use cloned voices for examples. Interestingly, cloning a voice-designed or a pre-made voice can actually improve its emotional range. Feel free to try this yourself.
Lowering stability can make the voice less steady, leading to stronger emotions, especially for intense or unusual performances like screaming. However, the AI is mainly great at narration, its main strength.
Rising anger, whispering to shouting, “No, you clearly don’t know who you’re talking to, so let me clue you in. I am not in danger, Skyler. I AM the danger. A guy opens his door and gets shot and you think that of me? No. I am the one who knocks!”
“I can’t believe this!” she said sadly, tears welling up in her eyes. “Is it really over?”
Screaming and in a low tone, “get the key your fool! We need to get our of here now!”
“Noooohuhuhu. I don’t want to!” she cried. “I want… to eat… my ice cream!” She sobbed uncontrollably.
“When you get to the gate, use the key! – and - make sure to not let… the.. demons… in!”
“Haha! That’s funny! I wish I would have thought of that. I guess it doesn’t make sense really. Hahahahha.” She giggled
Oi, mate, I tell ya, I was down in Southend last weekend, right? Sun was shining, everyone was out, and you won’t believe who I bumped into – our old mate Gaz from Romford! Ain’t seen him in donkey’s years!
“It sure does, Jackie… My mama always said: ‘In Carolina, the air’s so thick you can wear it!‘”
“If you want to introduce a pause ––- you can use dashes or … you can use an ellipsis”