Show HN: Run 25 Audio Models (TTS/STT/STS/Muisc) Locally in C++/GGML. No Python

Category: library

Tags: audio-inference, ggml, tts, stt, voice-cloning

Score: 7.5/10 (Innovation: 7, Technical: 8, Documentation: 7, Utility: 8)

audio.cpp is a high-performance C++ inference framework built on ggml that runs 25+ audio models (TTS, STT, music generation, voice cloning, etc.) locally without Python. It achieves significant speedups over Python reference implementations (up to 5x faster) and supports CUDA-optimized, portable, end-to-end audio workflows, making advanced AI audio models practical for production use.

Target audience: backend devs, ml engineers, audio engineers, devops

Repository: https://github.com/0xShug0/audio.cpp · C++ · NOASSERTION · 483 stars

View on Hacker News