Implement FreeVC

[erogol]

FreeVC implementation only with inference

#2438

From the original implementation -> https://github.com/OlaWod/FreeVC 👑@OlaWod

This PR allows voice cloning with any TTS model

from TTS.api import TTS
from TTS.utils.generic_utils import get_user_data_dir

print(get_user_data_dir("TTS"))

tts = TTS("tts_models/de/thorsten/tacotron2-DDC")

# regular TTS
wav = tts.tts_to_file("Wie sage ich auf Italienisch, dass ich dich liebe?", file_path="output_org.wav")

# the same model with voice cloning
wav = tts.tts_with_vc_to_file(
    "Wie sage ich auf Italienisch, dass ich dich liebe?",
    speaker_wav="target/speaker.wav",
    file_path="ouptut.wav"
)

You can also do voice conversion with the ```tts`` command. It will convert the voice of the source_wav to the speaker in the `target_wav`

tts --souce_wav original/audio/file.wav --target_wav target/audio/file.wav