Advanced-RVC-Inference and Multi-Model-RVC-Inference
These are ecosystem siblings where Multi-Model-RVC-Inference extends the architecture of Advanced-RVC-Inference by adding support for multiple simultaneous models and HuggingFace integration rather than replacing it.
About Advanced-RVC-Inference
ArkanDash/Advanced-RVC-Inference
Advanced RVC Inference for quicker and effortless model downloads
Provides both CLI and Gradio web interfaces for real-time voice conversion with batch processing capabilities, plus integrated audio source separation for isolating vocals from instrumental tracks. Built on PyTorch with optional CUDA acceleration, supporting pitch shifting and multiple inference modes through a unified command-line API. Incorporates dependencies from Applio, python-audio-separator, and NVIDIA's BigVGAN for comprehensive voice synthesis and audio manipulation workflows.
About Multi-Model-RVC-Inference
ArkanDash/Multi-Model-RVC-Inference
RVC Inference with multiple model and huggingface support
Supports V1 & V2 RVC models with integrated voice processing pipelines including Demucs source separation, TTS synthesis, and microphone input, deployable on HuggingFace Spaces with CPU-only constraints. Built on PyTorch with Hubert embeddings for voice conversion and optional RMVPE pitch extraction, providing both WebUI and command-line interfaces for flexible inference workflows.
Scores updated daily from GitHub, PyPI, and npm data. How scores work