search

vlomme / Multi-Tacotron-Voice-Cloning

Phoneme multilingual(Russian-English) voice cloning based on
https://github.com/CorentinJ/Real-Time-Voice-Cloning
Other
379 stars 97 forks source link

Training encoder #20

Open dtanalytic opened 3 years ago

dtanalytic commented 3 years ago

Thanks for work! Help me to train an encoder. How is it possible to add new custom voices to train datasets, or only fixed (like LibriSpeech: train-other-500, VoxCeleb1...) are available through the interface of commands: python encoder_preprocess.py and python encoder_train.py my_run /SV2TTS/encoder

If possible, than how i should keep files, in root data directory or subfolders, in what formats? I tried to add my voice to subfolder but got an error like:

"Python encoder_preprocess.py data Arguments: datasets_root: data out_dir: data/SV2TTS/encoder datasets: ['preprocess_voxforge'] skip_existing: False

Preprocessing preprocess_voxforge Couldn't find data/book, skipping this dataset"

I looked at the source and found that there are fixed funcs that preprocess different formats of train data (like preprocess22,preprocess44...) What do they mean? Maybe i should use one of them? Thank you.

vlomme commented 3 years ago

each dataset structure needs its own code. I have written functions for three options

dtanalytic commented 3 years ago

And what do this options mean (22,44,16)? Maybe i can use one of them?17.11.2020, 20:14, "Kramarenko Vladislav" notifications@github.com: each dataset structure needs its own code. I have written functions for three options

—You are receiving this because you authored the thread.Reply to this email directly, view it on GitHub, or unsubscribe.