It actually has nothing to do with that game. I didn’t even know it existed until you pointed it out! However, it is important to note that I’m not the first person to do this.
Well, I am using diphonic synthesis, which I do not believe has been done before, but monophonic synthesis has, and this is the same logic. So, as far as inspiration, you can thank previous developers such as @tyridge77, who made a fantastic demo of monophonic synthesis 6 years ago. In it, they challenge the viewer to make a diphonic version, which I took on as a challenge.
The purpose? The big one is accessibility. Text-to-speech isn’t something easily available in Roblox Studio, which is unfortunate. Text-to-speech can help with immersion, sure – such as when interacting with NPCs – but also some people may prefer it because they find it easier than reading.
Before you say it: Most people who deal with vision loss or reading disorders are already used to decades of robotic, unnatural speech synthesis. And, trust me, if I could get a more realistic method, I would.
For me, I am really looking forward to generating some terrifying voices for any psychological horror games that I am either currently working on or will work on in the future.
See, I made a tutorial on AI text generation recently. If I’m going to use that, the audios for text-to-speech certainly cannot be pre-recorded because AI is unpredictable.
This also goes for text that is usually predictable but may change slightly. For example, if you want a character in your game to refer to someone by their username, you cannot prerecord that dialogue, either.
For other uses, this is a novelty, or maybe a luxury to some. However, the topic has gained so much traction over the years that I am willing to make something and put it out there when I can.
When will that be? As soon as Roblox lets me distribute the audio in the store. It’s just one audio.
Thank you for your reply! :)
