From 7303033dc6267128c148610de0f1b17d59e77b02 Mon Sep 17 00:00:00 2001 From: Martin Warnaar Date: Tue, 4 Jul 2023 14:30:31 +0200 Subject: [PATCH] readme : better wording (#1064) --- README.md | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/README.md b/README.md index 58a894f330c..a429d9580a4 100644 --- a/README.md +++ b/README.md @@ -313,7 +313,7 @@ For more information about the Core ML implementation please refer to PR [#566]( ## NVIDIA GPU support via cuBLAS -With NVIDIA cards, the Encoder processing can be offloaded to the GPU to a large extend through cuBLAS. +With NVIDIA cards the Encoder processing can to a large extent be offloaded to the GPU through cuBLAS. First, make sure you have installed `cuda`: https://developer.nvidia.com/cuda-downloads Now build `whisper.cpp` with cuBLAS support: @@ -325,7 +325,7 @@ WHISPER_CUBLAS=1 make -j ## OpenCL GPU support via CLBlast -For cards and integrated GPUs that support OpenCL, the Encoder processing can be largely offloaded to the GPU through CLBlast. This is especially useful for users with AMD APU's or low end devices for up to ~2x speedup. +For cards and integrated GPUs that support OpenCL, the Encoder processing can be largely offloaded to the GPU through CLBlast. This is especially useful for users with AMD APUs or low end devices for up to ~2x speedup. First, make sure you have installed `CLBlast` for your OS or Distribution: https://github.com/CNugteren/CLBlast