Setup Environment

(Tested on a Quadro RTX 5000 with NVIDIA-SMI Driver Version: 535.104.05, CUDA Version: 12.2 on a UBUNTU 22.04)

Build Dockerfile
```
 docker build -t whispervits-svc .
```
Enter Docker container
Download the Timbre Encoder: Speaker-Encoder by @mueller91, put best_model.pth.tar into speaker_pretrain/.
Download whisper model whisper-large-v2. Make sure to download large-v2.pt，put it into whisper_pretrain/.
Download hubert_soft model，put hubert-soft-0d54a1f4.pt into hubert_pretrain/.
Download pitch extractor crepe full，put full.pth into crepe/assets.

Note: crepe full.pth is 84.9 MB, not 6kb
Download trained model lesd5_100.pretrain.pth, and put it into vits_pretrain/.
Make sure you have downloaded the wav_spk_1 folder from the Benchmarking-SGDD repository. Then, run the script.

python convert-TWH-spk1.py /path/to/wav_spk_1

The output will be a folder containing all conversions used on the evaluation. The same that is found on this google drive.

Name		Name	Last commit message	Last commit date
Latest commit History 14 Commits
configs		configs
crepe		crepe
data_svc_5/singer		data_svc_5/singer
feature_retrieval		feature_retrieval
files		files
hubert		hubert
hubert_pretrain		hubert_pretrain
pitch		pitch
speaker		speaker
speaker_pretrain		speaker_pretrain
vad		vad
vits		vits
vits_decoder		vits_decoder
vits_extend		vits_extend
vits_pretrain		vits_pretrain
whisper		whisper
whisper_pretrain		whisper_pretrain
Dockerfile		Dockerfile
LICENSE		LICENSE
README.md		README.md
build_image.sh		build_image.sh
convert-TWH-spk1.py		convert-TWH-spk1.py
environment.yml		environment.yml
recod_container.sh		recod_container.sh
requirements.txt		requirements.txt
svc_eva.py		svc_eva.py
svc_export.py		svc_export.py
svc_inference.py		svc_inference.py
svc_inference_batch.py		svc_inference_batch.py
svc_inference_post.py		svc_inference_post.py
svc_inference_shift.py		svc_inference_shift.py
svc_merge.py		svc_merge.py
svc_preprocessing.py		svc_preprocessing.py
svc_train_retrieval.py		svc_train_retrieval.py
svc_trainer.py		svc_trainer.py

Provide feedback