I have some audio demos in the GitHub you can view and a gui for everything No GPU is needed for any of this I've integrated GPT-4 for human level (who said what) but costs money for their api use, BOOKLP for FREE and LOCAL (who said what) Llama2 13B (who said what) failed is just really bad under experimental folder Balacoon (super duper fast local audio generation) Tortoise (super high quality voice synthesis super sloW) Bark(super slow multi lingual model high quality audio) Google colab versions of everything if you wanna do that. Give me your thoughts! This has been my dream for a while and I decided to try remaking the whole thing from the ground up. still work in progress progress on it I'll update the readme to any suggestions and such. [link] [comments] |