Ggml-medium.bin
-osrt : Output the transcription directly into a SubRip ( .srt ) subtitle file, perfect for video editing.
./build/bin/whisper-cli -m models/ggml-medium.bin -f samples/my_audio_file.wav Use code with caution. 3. Output Formats ggml-medium.bin
: It works natively across Intel, AMD, ARM, and Apple Silicon architectures. -osrt : Output the transcription directly into a SubRip (
This script will fetch the ggml-medium.bin file and place it securely into your ./models directory. Step 3: Build the Main Executable ggml-medium.bin
To understand what makes ggml-medium.bin unique, it is necessary to break down its two foundational components: OpenAI's Whisper and Georgi Gerganov's GGML library.
Apple M1, M2, or M3 chips run this model exceptionally well by utilizing the Apple Neural Engine and unified memory via whisper.cpp .
It handles multilingual transcription and translation tasks with high precision.