Description

Last Update : <t:1703022101:R> - Model URL Version RVC V2.0 - Pitch Extraction Algorithm RMVPE - Epochs - Steps 300 - 13.5 K - Dataset ~ 00:19:37 - Recommended Usage Speech - Search Feature Ratio 0.75 ( You can lower it if you use a language other than Japanese, although in some cases this is not required ! ) - Pitch Man = -20 > 0 (Depend base voice) - Woman = Man Adjustement - 12 (Depend base voice) - You can adjust if you found an better result Previews Preview_TTS.wav *Provide* : ElevenLabs - Daniel Voice [JP] - *Contains External Effects* : No - *Pitch* 8 - *Feature Ratio* : 0.75 - Preview_Cover.wav *Caution* : I've provided an example of a cover, but as explained above, the model isn't really designed for singing ! - *Contains External Effects* : Yes - *Pitch* 20 - *Feature Ratio* : 0.75