kigner/audio.cpp-webui

audio.cpp with a full-task WebUI - pure C++ audio-model inference engine powered by ggml. TTS, ASR/STT, VAD, voice conversion, speaker diarization, music generation. No Python dependency.

What it solves

This project provides a high-performance, portable C++ runtime for running various audio AI models locally. It eliminates the need for complex Python environments and dependency conflicts by providing a shared native runtime that works across Windows, Linux, and macOS, supporting NVIDIA, AMD, and Apple Silicon hardware.

How it works

Built on top of ggml, the framework provides optimized execution paths for audio models. It supports GGUF loading and quantization (e.g., Q8), which reduces VRAM usage and increases inference speed. The system includes a Gradio-based WebUI for model management and execution, as well as CLI and server entry points. It also features built-in audio utilities for denoising, resampling, and enhancement.

Who it’s for

Developers and users who want to run state-of-the-art audio models (TTS, ASR, voice cloning) locally with high efficiency and minimal setup overhead, especially those targeting production-style end-to-end execution rather than simple demos.

Highlights

  • Broad Model Support: Supports over 40 model families across tasks like Text-to-Speech (TTS), Automatic Speech Recognition (ASR), voice cloning, source separation, and music generation.
  • Extreme Performance: Significant speedups over Python reference paths, with some models running up to 200x real-time on CUDA.
  • Hardware Portability: Native support for CUDA, HIP/ROCm, Vulkan, Metal, and CPU backends.
  • GGUF Integration: Extensive use of GGUF for efficient model loading and reduced memory footprints.
  • Integrated WebUI: A user-friendly interface for managing downloads and running audio workflows without CLI commands.

Related

  • Project
    0xShug0/audio.cpp

    A high-performance C++ audio inference framework built on ggml that provides a portable, native runtime for running diverse local audio models including TTS, ASR, and voice conversion.

  • Project
    foldl/chatllm.cpp

    A C++ LLM inference engine based on ggml that enables memory-efficient local execution of multimodal models with RAG and quantization support.

  • Project
    NVIDIA/TensorRT-Edge-LLM

    A high-performance C++ inference runtime for deploying LLMs and multimodal models on NVIDIA Jetson, DRIVE, and DGX Spark edge platforms.

  • Project
    handy-computer/transcribe.cpp

    transcribe.cpp is a native C/C++ library for fast speech‑to‑text inference. It runs dozens of modern STT models (Parakeet, Whisper, Qwen3‑ASR, Voxtral, etc.) using the GGUF format on ggml, with Metal, Vulkan, CUDA, ROCm, or a tinyBLAS‑accelerated CPU backend. The repo provides a CLI, a quantization tool, and official Python, TypeScript, Rust, and Swift bindings, all under an MIT license.

  • Project
    jatinchowdhury18/RTNeural

    A lightweight C++ neural network inferencing engine optimized for real-time systems and audio processing.