Scarlett Johansson

Link: https://huggingface.co/datasets/fckmylife/scarlett_johansson/resolve/main/Scarlett_Johansson.zip Dataset length: 23 minutes Epochs: 300 RVC v2 ℹ️ trained on two different source: - The Howard Stern Show Podcast (2017) - Her (2013) Movie There are two different index files: ``added_IVF1651_Flat_nprobe_1_Scarlett-Johansson_v2.index`` (Podcast speech / normal pitch) ``added_IVF828_Flat_nprobe_1_Scarlett_Johansson_HER_v2.index`` ('Her' AI voice pitch ) It's far from perfect, but it will do the job as I primaly use as my home tts AI voice. If I get a better source, I'll update the model here. Works best with female voice as source. credits <@716624198688178187> ```🔆 Check out my other posts of unique voice models with example audio files. If you're interested in a personalized high quality voice model, just shoot me a DM!```

tts test ('her' .index) + random voicemail sample (real female voice)

I'm not sure what do you mean, but I'm using Retrieval based Voice Conversion (RVC) WebUI locally to train these models.

which TTS do you use

For the demos, mostly edge-tts Py

is this no longer available?

Does this include her singing voice?

DEMO: whyp.it/tracks/110888/scarlett-johansson-rvc-v2-300-epochs

How can i use this model locally for local voice assistant?