whisper.cpp
whisper.cpp copied to clipboard
ggerganov
Port of OpenAI's Whisper model in C/C++
Needed for mingw32 build to successfully install the dlls in the correct location. I realise the default setting in the CMakefile disables the shared library build for mingw32, but when...
Hi, I added a class to allow flexible loading of whisper model from any source. Please note that I made a breaking change (renaming `whisper_init` to `whisper_init_from_file`) and didn't test...
On https://github.com/ggerganov/whisper.cpp/tree/master/examples/stream page the "Building" section mentions you just need to have "libsdl2-dev" package installed, but in fact you also need "ffmpeg". Without it, the example was failing for me...
- update from 3.0 (from 2014) to 3.19 (from 2020) - move some global setting onto the targets (through a cmake include) Right now it's not possible to include whisper.cpp...
Here is a first attempt to clang-format as close to the existing format as possible ([#332](https://github.com/ggerganov/whisper.cpp/issues/332)). I formatted the whisper and ggml files to show what it does. If there's...
Running the binary on the Mac build: ```yml - name: Test run: | bash models/download-ggml-model.sh tiny ./main -m models/ggml-tiny.bin -f samples/jfk.wav ``` ...raises the following error: ```shell 4062 Illegal instruction:...
Hello there, After doing some smoke tests of whisper.cpp utilizing ./main (all of that was working just perfectly with different language models) I moved to "stream" example - https://github.com/ggerganov/whisper.cpp/tree/master/examples/stream The...
With the FP32 base model and this patch set, the jfk example takes about 3.2 seconds to transcribe. This is another data point for #300, and it is about one...
I changed GGML_F16_VEC_LOAD and GGML_F16_VEC_STORE to pass the index as a separate argument. I haven't tested those changes on the other architectures. Will they be tested automatically before the merge?