whisper.cpp
whisper.cpp copied to clipboard
Port of OpenAI's Whisper model in C/C++
I am interested in using the streaming tool for a very specialized context (radiology dictation), which uses an esoteric and relatively restricted vocabulary (i.e. high likelihood of words like "hyperpneumatization"...
ggml.c fails to compile on Debian ppc64le, GCC 11, with: ggml.c:135:10: fatal error: immintrin.h: No such file or directory 135 | #include If I comment out that line, the build...
ref #220 WIP IN PROGRESS
The timestamp of whisper is not very accurate. The following is the comparison between Microsoft Cognitive Services Speech and whisper. ``` 1 00:00:00,120 --> 00:00:01,379 (Microsoft) [00:00:00.000 --> 00:00:02.000] (whisper)...
This makes it more organised and easier to work with afterwards in my opinion, and it also keeps it in parity with the [original implementation](https://github.com/openai/whisper/pull/101).
'-Ofast' and '-march=native' cause 2x-speedup in machines with SSE (but no AVX) instructions. Should help other platforms, too.
I have a Celeron-equipped machine; which has SSE, but no AVX instructions. I just wanted to let you know that processing the JFK sample... - Takes **73 seconds** with the...
I have a file which according to `ffprobe` is 6m59.53s. The end of the output looks like this: ``` [00:06:41.360 --> 00:06:47.840] In fact, rather than a protection of the...
Is there any parameter that needs to be added into the implementation like in https://github.com/openai/whisper/tree/main/whisper/assets/multilingual ? I've tested all models and found the inferenced results are different compared to using...
ref #129 The performance is comparable to when running `whisper.cpp` in the browser via WASM. The API is currently very rudimentary: https://github.com/ggerganov/whisper.cpp/blob/npm/bindings/javascript/emscripten.cpp I am hoping that there will be interest...