# Before Start > - This model is quite **special**, as it's based on an interview he did while **speaking French**! So I agree that it's not natively French, but I wanted to see if the result was clean. This makes it possible to create a "variant" to avoid posting a model that's already been posted before ! > - Don't forget it's a test model version, i upload all models when the 48k final version of model has released ! # Last Update : <t:1704994153:R> > - **Model URL :** > - https://huggingface.co/rayzox57/RobertOppenheimer_RVC/resolve/main/RobertOppenheimer_French_Speech_v2_Ov2_300e.zip > - **Version :** > - <a:firev2:1167361499149381662> RVC V2.0 - Ov2 Super > - **Pitch Extraction Algorithm :** > - RMVPE > - **Epochs - Steps :** > - 300 - 4.2k > - **Dataset :** > - ~ 00:06:12 [ 1958 : Entretien avec Oppenheimer, le père de la bombe atomique | Archive INA ] > - It's an old archive, It's quite normal if there are small artifacts, or small robotic passages (cleaning the archive must have contributed to this). > - **Recommended Usage :** > - Speech > - **Search Feature Ratio :** > - 0.75 > - **Pitch :** > - ( 0 = Man ) > - You can adjust if you find better version # Previews : > - **Preview_TTS_VF.wav :** > - *Provide* : TTS - ElevenLabs v1 - Josh FR > - *Contains External Effects* : No > - *Pitch* : 0 > - *Feature Ratio* : 0.75 > > - **Preview_TTS_En.wav :** > - *Provide* : TTS - ElevenLabs v1 - Josh EN > - *Contains External Effects* : No > - *Pitch* : 0 > - *Feature Ratio* : 0.75 > - *Info* : Just for fun

Reference : https://www.youtube.com/watch?v=5E0r-ssYI94