Soul-AILab/SoulX-Singer

add fp16 suport for svc

optimize gpu memory usage

feat: add svc inference code and webui

Update webui.py

Update webui.py

Update webui.py

Update webui.py

Upload 3 files

Update webui.py

Update webui.py

Update webui.py

Upload 2 files

Update webui.py

Update webui.py

Update webui.py

Fix LlamaAttention compatibility with transformers 4.53.0

Fix NeMo CUDA error 35 on ZeroGPU by disabling CUDA graphs in decoding config

Make _get_device() ZeroGPU-aware so models target CUDA

Update requirements.txt

Disable NeMo CUDA Graphs to fix CUDA failure 35 on RNNT decoding

Fix ZeroGPU nested @spaces.GPU causing CUDA failure 35

Update requirements.txt

Update requirements.txt

Update preprocess/tools/lyric_transcription.py

Update preprocess/tools/g2p.py

Update requirements.txt

Update requirements.txt

Update requirements.txt

Update app.py

Update requirements.txt

Update requirements.txt

Update webui.py

Fix model download reliability and silent import failures

Update requirements.txt

Upload webui.py with huggingface_hub

Upload app.py with huggingface_hub

Upload requirements.txt with huggingface_hub

Upload README.md with huggingface_hub

Upload requirements.txt with huggingface_hub

Upload requirements.txt with huggingface_hub

Upload folder using huggingface_hub

initial commit