whisper.cpp
whisper.cpp copied to clipboard
ggerganov
Port of OpenAI's Whisper model in C/C++
 I am using whisper.cpp (most recent commit [a56f435](https://github.com/ggerganov/whisper.cpp/commit/a56f435fd475afd7edf02bfbf9f8c77f527198c2)) and the large-v3 model and I am getting the impression that the phenomenon of whisper getting stuck in...
With the recent git update, the `server --convert` option no longer works properly all the time. After the second and subsequent access to the server, it responds with Enter command:...
I've recently compiled the 1.5.4 library for with CUBLAS and having an issue where running multiple whisper_full_with_state()'s. I did not previously have this issue with the 1.5.1 library. I re-compiled...
I have been able to appreciate significant performance increases utilizing Openvino for the models that I can covert, but my laptop does not seem to have enough ram (16 GB)...
I'm a bit stuck and maybe someone can help me. Running whisper large-v3 model via whisper.cpp is significantly more performant than running it through python, VRAM wise and time wise....
Whenever I try to speak in any language other than English, it either converts it in english, or the output reads - "Speaking in foreign lanuage". My use case mostly...
- We need to ignore Apple platforms frameworks (Accelerate and Metal) when building the project on Linux - We need to define `_GNU_SOURCE` for certains function declaration (in `sched.h`)
In ggml_backend_cuda_buffer_type function you have static array ggml_backend_cuda_buffer_types[GGML_CUDA_MAX_DEVICES] which you fill with allocated memory in "context" field which is not freed.
We face the next issue while creating each backend as separate loaded DLL. In your code "ggml-backend.c" there are two methods for backend buffer creation: ggml_backend_buft_alloc_buffer - this method call...
It's possible to fine-tune models to be able to use audio_ctx more freely, without affecting their knowledge too much. Example with default settings (notice the ~3x speed difference): ``` $...
