Hi. What would be the best way to create speech for a created language based on English phonemes? I’ve had some generators for creating written languages with pronunciation guides for years but no software to speak the languages based on the pronunciation. For a future project I’m thinking of creating one preferably using dictionaries for the spoken phoneme and it’s written form in the generated language. Any recommendations?
FYI the generators are pen and paper tables, not software but I can code the written form easy enough.
Use Hangul (Korean). It is a phonetic alphabet created in the 1400s that is still in use today. You will need to ap your made up language to it, but in doing so, you can get something to pronounce it pretty easily. Each character, or combination of characters being a sound, means you can create any sound you need in writing.
Thanks I’ll check it out but it may not work as the pronunciation is based on English. There seems to be several options as far as the Godot part goes.
I don’t think you understand what a phonetic alphabet is. You can write any language in Hangul, including English. That’s the point.
This sounds like a wast topic, as this can be taken in many directions depending on how you see it in your mind. A funny, toy example of the synthetic speech is the “Animalese” from the animal crossing, which I think is well documented online.
It’s how to code the speech that I’m interested in.
I can also use the standard phonetic alphabet. Whichever I use I still need to code it so it sounds realistic. It’s the best approach to coding it I’m after. Preferably putting the words together from the sounds at runtime.
That’s not a thing.
Godot has text to speech functionality built-in.
Text to speech but not necessarily the individual sounds that make up the words. How is it not a thing? Unicode supports the international phonetic alphabet. Are you suggesting using text to speech with Hangul? This needs to work with phonemes rather than complete words.
Might use an external API for the speech in which case Godot would be writing the Unicode and playing back the audio from the API.
Looking at Hangul. It’s definitely not going to get all the pronunciation I need. Unfortunately TTS Maker looked promising at first but doesn’t seem to accept custom words even when spelt phonetically.