whisper.cpp
whisper.cpp, fast, on-device speech-to-text powered by OpenAI's Whisper, in plain C/C++. No cloud calls.
Release history
whisper.cpp 1.9.5
Whisper.cpp 1.9.5 updates to ggml v0.26.0, bringing GPU optimizations across CUDA, Vulkan, Metal, and WebGPU for faster speech-to-text inference
whisper.cpp 1.9.4
Whisper.cpp 1.9.4 fixes encoder callback timing, decoder state management, and syncs GGML 0.23.0 with multi-platform backend improvements
whisper.cpp 1.9.3
Whisper.cpp 1.9.3 syncs GGML backend with performance optimizations across CUDA, Metal, SYCL, and other accelerators including quantization support and memory improvements
whisper.cpp 1.9.2
Syncs with ggml; Includes Parakeet in build-xcframework.sh and fixes pkgconfig configuration
whisper.cpp 1.9.1
Release: v1.9.1
whisper.cpp 1.9.0
Adds NVIDIA Parakeet model support. Check changelog if running Ruby bindings
whisper.cpp 1.8.7
Drops hardcoded /usr/lib assumption; Relevant if installing to non-standard prefixes
whisper.cpp 1.8.6
Whisper.cpp 1.8.6 refactors ffmpeg transcoding logic and fixes CI header path resolution
whisper.cpp 1.8.5
Whisper.cpp 1.8.5 fixes server parameter handling, VAD streaming, Ruby binding memory leaks, and timestamp accuracy