whisper.cpp
C/C++ port of OpenAI Whisper for offline speech-to-text inference.
Tested on macOS
Use it
macOS arm64
brew install whisper-cpp
brew install whisper-cpp
Check that it works
whisper-cli --version 2>&1The output should contain “whisper.cpp version: 1.9.5”.
First thing to try
mkdir -p /tmp/whisper_test/models /tmp/whisper_test/samples
# Download jfk.wav sample
curl -fL -o /tmp/whisper_test/samples/jfk.wav https://github.com/ggml-org/whisper.cpp/raw/master/samples/jfk.wav 2>&1 | tail -n 5
ls -lh /tmp/whisper_test/samples/jfk.wavcurl -fL -o /tmp/whisper_test/models/ggml-tiny.en.bin https://huggingface.co/ggerganov/whisper.cpp/resolve/main/ggml-tiny.en.bin 2>&1 | tail -n 5
ls -lh /tmp/whisper_test/models/whisper-cli -m /tmp/whisper_test/models/ggml-tiny.en.bin -f /tmp/whisper_test/samples/jfk.wav 2>&1What it does
- Transcribe audio files to textTested
- Transcribe microphone audio in real timeNot tested
- Quantize Whisper models to integersNot tested
- Highlight transcription confidence with colorsNot shown working
How far pickfit got
macOS arm64
- Installs
- Starts
- 1 of 4 features work
- Works offline
How pickfit tested it
Each run starts from a clean machine and follows the steps above. How pickfit verifies
macOS arm64
Its main function works
- Replayed from scratch
- Works offline
- Checked on 4 Nov 2026
- Version 1.9.5
Differences the test machine hit
- README: ./build/bin/whisper-cli -f samples/jfk.wav with model in models/
In the test: brew install places binaries as whisper-cli on PATH; sample wav and ggml models are not bundled and must be downloaded separately (e.g., curl jfk.wav and ggml-tiny.en.bin from Hugging Face huggingface.co/ggerganov/whisper.cpp)