Does it support more langages apart from english?
Does it support more langages apart from english?
it was also trained in French, Spanish, etc. languages. You should specify this in the description: "French voice, french accent, etc."
I’m trying in the space here, because I’m on my phone, but even using only “Spanish voice, Spanish accent” I get only results like an American person trying to read Spanish, not as an speaking speaker at all, am I doing something wrong?
Is there some specific speaker trained in languages?
Realmente lo único que he encontrado que funciona de forma decente es Bark, aún así tienes que usar despues RVC para conseguir buen resultado y no es 100% estable ni controlable, pero da resultados decentes.
Para Bark utiliza Bark Infinity para poder entrenar voces, no sirve para clonar por que no clona aunque se supone que es para eso, pero te ayuda a definir acentos.
Si la licencia no es un problema para ti, entonces sin NINGUNA duda, Coqui o Tortoise con XTTSv2 :)
¡Espero que te sea de ayuda!
Guys I've just created a tutorial on how to fine-tune a text-to-speech model in any language. If you can find a dataset consists of Spanish language, you can easily create a model.
The tutorial video:
https://www.youtube.com/watch?v=TZIBQ24UCgA
Training codes:
https://github.com/emirhanbilgic/Turkish-TTS
Dataset I used:
https://huggingface.co/datasets/erenfazlioglu/turkishvoicedataset
Hugging face demo to try:
https://huggingface.co/spaces/emirhanbilgic/Text-to-speech-Turkish
@emirhanbilgic thanks, I'll take a look!
How much GPU time is needed to do so?
i mean can it be done with an RTX4090 for example?
Thanks!
Hey @juang3d ! You can easily do it with RTX 4090. I did it with T4 in just 20 mins of training! It would take ~13 mins for RTX 4090 :)