Genshin Impact Voice Models + (Added Zhongli 80 Min ripped dataset)

Description

Genshin Voice Models (RVC) - Gorou, Xiao, Lisa, Childe (Tartaglia), Zhongli English VA's. Feel free to request for others. - 500 epochs, locally trained - Trained with ripped voice lines from the game with CREPE at 32 hops (Mangio RVC Beta). - All samples normalized and denoised professionally with Waves/iZotope plugins using actual noise samples before training. - Trained with pitch, should work well for both spoken word and songs. - Would recommend inferring with CREPE, experiment with hop length and feature retrieval. 128-192 with 0.4-0.8 usually sounds good. - Examples attached were just recorded with a mediocre headset mic, no mixing at all. I also have a low pitch voice and sound nothing like these characters. Inferred on someone with a closer voice to the target model + better mic would give much better results. Please credit me if used publicly anywhere such as social media etc. Discord: BigSoulja#8888 TikTok: @souljavr YouTube: You can credit any one of the above handles, does not need to be all of them. Xiao (500 Epoch RVC V1): Childe (500 Epoch 32 Hop Length RVC V1): Gorou (500 Epoch 32 Hop Length RVC V1): Lisa (500 Epoch 32 Hop Length RVC V1): Zhongli (750 Epoch 16 Hop Length RVC V2, Huge 80 Min Ripped Dataset): Keqing (300 Epoch 16 Hop Length RVC V2): Will be adding Genshin AI models to my huggingface repository too: