Do I know anyone who can help me to make a custom text-to-speech voice? There are proprietary services that allow you to do this, but they rely on machine learning APIs that are quite expensive to run, and I wouldn't independently own the result. I would maybe like to be able to license a simulated version of my own voice for other trans artists and allies to use for free, to make their work more accessible.
Not that I have particularly great voice, nor am I in any way trained as voice talent, but when I'm looking for text-to-speech voices for my own work, I really struggle with every voice being heavily gendered and none of them being Northern English. My voice has very little else going for it, but it is at least androgynous and Northern!
I'm starting to think http://mycroft-ai.gitbook.io might be the direction to go in: similar technology to the big tech companies, but it's open source?
@zoy if you're looking for a state of the art, natural sounding voice, my understanding is that the commercial ML stuff like Google's WaveNet is the best we have at the moment.
For something less fancy, you could always go with the older school strategy of recording a wide array of phonetics and then stringing it together correctly. Way more time consuming, but it makes it a very auditable process.
@zoy Conversation with my colleague Lou Carroll who's pretty knowledgeable about singing voice synthesis
@evan thanks so much - janky is maybe fine, if it sounds artificial that's great. But yeah I think I need something with a clear roadmap.
Revel in the marvels of the universe. We are a collective of forward-thinking individuals who strive to better ourselves and our surroundings through constant creation. We express ourselves through music, art, games, and writing. We also put great value in play. A warm welcome to any like-minded people who feel these ideals resonate with them.