Description

Before Start The model seems a bit unstable. I really tried to do my best, if one day I have to redo a version I will take more time to add much more dataset, while improving the dataset as much as possible (even if for the moment I did my best). - Mr. Linea is a character who uses a lot of stammering to make him more comical, and this is what makes the model a little less qualitative, as it's hard for the model to understand how these stammering's are useful when used. - As the animation was made in the 70's, it's important to note that the sound quality isn't the best either. Last Update : <t:1703878612:R> - Model URL Version RVC V2.0 - Pitch Extraction Algorithm RMVPE - Epochs - Steps 800- 15.2k - Dataset ~ 00:03:43 ( Some Episodes ~10 ) - Recommended Usage Cover - Search Feature Ratio 0.0 ( I recommend you to set close 0.0 to reduce as maximum the noise ) - Pitch Logic Pitch ( +8 = Man ) (depend of man voice used) - You can adjust if you found an better result Previews Preview_TTS.wav *Provide* : ElevenLabs V1 - Italian - Patrick - Stability: 100% - Clarity + Similarity : 100% - *Contains External Effects* : No - *Pitch* : +8 - *Feature Ratio* : 0.0