whisper.cpp

48b18f9e - ggml : prevent integer overflow in gguf tensor size calculation (llama/14595)