Almost instant voice cloning
This project was already on hf and was improved by me for colab and local
Coming soon..
- UI
. This is the colab for tts, in rot13 and base64 (Low risk of ban or almost totally cancelled) (guide along with local version will be released soon)
Check the Hugging Face Space 🤗
-
Run "Clone and Install" and wait a few minutes, when a green check appears, it was successful
-
Run "Run" and when appears 2 link blue, prizes the second link "public link"
-
after opening the public link, follow this guide
- in "text" put your text
- in "URL" (10 seconds is fine)!" put your link that you uploaded to hugginface, with your dataset, mp3,mp4,flac,wav
- or upload your file in "input file" ps: you have to reload the audio every time you do a conversion :( we are fixing it
- in "Language" you can select the language of the text written before, or the language you want your character to speak
- now press submit and wait a few seconds, usually with a text about 40 characters long it takes 7/10 seconds in colab, 15/17/20 seconds in huggingface.co
- now your audio will have appeared in "output" which you can either listen to from gradio or install with the arrow