whisper.cpp and whisper_ros
The latter is a ROS 2 package that integrates the former, Whisper.cpp, for speech-to-text functionality within a robotics operating system environment, making them complements where one builds upon the other.
About whisper.cpp
ggml-org/whisper.cpp
Port of OpenAI's Whisper model in C/C++
Optimized for resource-constrained environments through integer quantization, mixed-precision inference (F16/F32), and zero runtime memory allocations, enabling on-device ASR on mobile and embedded platforms. Leverages the GGML inference library with multi-platform GPU acceleration via Metal, Vulkan, CUDA, and Core ML, alongside CPU-optimized SIMD paths for ARM NEON, AVX, and POWER VSX architectures. Provides a minimal C API and supports deployment across iOS, Android, WebAssembly, Raspberry Pi, and standard desktop/server platforms.
About whisper_ros
mgonzs13/whisper_ros
Speech-to-Text based on SileroVAD + whisper.cpp (GGML Whisper) for ROS 2
Integrates whisper.cpp (GGML format) with ROS 2's audio_common for unified audio I/O, while SileroVAD preprocesses streams to reduce transcription overhead by detecting voice activity. Supports hardware acceleration via CUDA for both Whisper inference and SileroVAD ONNX runtime, with configurable model paths, sampling parameters, and OpenVINO device selection for cross-platform deployment.
Related comparisons
Scores updated daily from GitHub, PyPI, and npm data. How scores work