public_git/whisper.cpp
1
0
Fork 0
mirror of https://github.com/ggml-org/whisper.cpp.git synced 2026-09-30 19:41:14 +02:00
Code Issues Packages Projects Releases Wiki Activity

whisper : disable CUDA mel + fix FFMPEG

Browse Source
This commit is contained in:
Georgi Gerganov
2024-06-26 20:11:38 +03:00
parent 3efedb9511
commit dc8cc2dd6f
6 changed files with 68 additions and 41 deletions
Download Patch File Download Diff File Expand all files Collapse all files
examples/CMakeLists.txt
+30 -5
View File
@@ -11,7 +11,7 @@ if (WHISPER_SDL2)
string(STRIP "${SDL2_LIBRARIES}" SDL2_LIBRARIES)
message(STATUS "SDL2_INCLUDE_DIRS = ${SDL2_INCLUDE_DIRS}")
message(STATUS "SDL2_LIBRARIES = ${SDL2_LIBRARIES}")
message(STATUS "SDL2_LIBRARIES = ${SDL2_LIBRARIES}")
endif()
if (WHISPER_CLBLAST)
@@ -22,10 +22,35 @@ endif()
set(TARGET common)
unset(COMMON_EXTRA_LIBS)
if (WHISPER_FFMPEG)
# As of cmake 3.27, there is no official cmake support for FindFFmpeg.
# Consequnelty we added a FindFFmpeg.cmake script the cmake subfolder:
# whisper.cpp does not need the full ffmpeg libs, just AVFORMAT AVCODEC AVUTIL SWRESAMPLE
# libswresample performs highly optimized audio resampling, rematrixing and sample format conversion operations
# libavcodec provides a generic encoding/decoding framework and contains multiple decoders and encoders for audio, video and subtitle streams, and several bitstream filters.
# libavformat provides a generic framework for multiplexing and demultiplexing (muxing and demuxing) audio, video and subtitle streams.
find_package(FFmpeg REQUIRED)
if (NOT ${FFMPEG_FOUND})
message(FATAL_ERROR "Cannot find ffmpeg libs/headers")
endif()
message(STATUS "Found ffmpeg libs: ${FFMPEG_LIBRARIES}")
message(STATUS "Found ffmpeg headers in: ${FFMPEG_INCLUDE_DIRS}")
message(STATUS "ffmpeg definitions: ${FFMPEG_DEFINITIONS}")
message(STATUS "Found avformat ${AVFORMAT_VERSION}")
include_directories(${FFMPEG_INCLUDE_DIRS})
add_compile_definitions(WHISPER_FFMPEG)
list(APPEND COMMON_EXTRA_LIBS ${FFMPEG_LIBRARIES})
set(COMMON_SOURCES_FFMPEG ffmpeg-transcode.cpp)
endif()
add_library(${TARGET} STATIC
common.h
common.cpp
@@ -38,7 +63,7 @@ add_library(${TARGET} STATIC
include(DefaultTargetOptions)
target_link_libraries(${TARGET} PRIVATE whisper)
target_link_libraries(${TARGET} PRIVATE whisper ${COMMON_EXTRA_LIBS})
set_target_properties(${TARGET} PROPERTIES POSITION_INDEPENDENT_CODE ON)
set_target_properties(${TARGET} PROPERTIES FOLDER "libs")
@@ -55,8 +80,8 @@ if (WHISPER_SDL2)
include(DefaultTargetOptions)
target_include_directories(${TARGET} PUBLIC ${SDL2_INCLUDE_DIRS})
target_link_libraries(${TARGET} PRIVATE ${SDL2_LIBRARIES})
target_include_directories(${TARGET} PUBLIC ${SDL2_INCLUDE_DIRS})
target_link_libraries (${TARGET} PRIVATE ${SDL2_LIBRARIES})
set_target_properties(${TARGET} PROPERTIES POSITION_INDEPENDENT_CODE ON)
set_target_properties(${TARGET} PROPERTIES FOLDER "libs")
@@ -108,7 +133,7 @@ if (WHISPER_SDL2)
set_target_properties(talk-llama PROPERTIES FOLDER "examples")
add_subdirectory(lsp)
set_target_properties(lsp PROPERTIES FOLDER "examples")
if (LLAMA_SYCL)
if (GGML_SYCL)
add_subdirectory(sycl)
set_target_properties(sycl PROPERTIES FOLDER "examples")
endif()
examples/common.cpp
+3 -3
View File
@@ -30,7 +30,7 @@ extern bool ffmpeg_decode_audio(const std::string & ifname, std::vector<uint8_t>
#endif
// Function to check if the next argument exists
std::string get_next_arg(int& i, int argc, char** argv, const std::string& flag, gpt_params& params) {
static std::string get_next_arg(int& i, int argc, char** argv, const std::string& flag, gpt_params& params) {
if (i + 1 < argc && argv[i + 1][0] != '-') {
return argv[++i];
} else {
@@ -346,7 +346,7 @@ std::vector<gpt_vocab::id> gpt_tokenize(const gpt_vocab & vocab, const std::stri
return tokens;
}
std::vector<gpt_vocab::id> parse_tokens_from_string(const std::string& input, char delimiter) {
static std::vector<gpt_vocab::id> parse_tokens_from_string(const std::string& input, char delimiter) {
std::vector<gpt_vocab::id> output;
std::stringstream ss(input);
std::string token;
@@ -358,7 +358,7 @@ std::vector<gpt_vocab::id> parse_tokens_from_string(const std::string& input, ch
return output;
}
std::map<std::string, std::vector<gpt_vocab::id>> extract_tests_from_file(const std::string & fpath_test){
static std::map<std::string, std::vector<gpt_vocab::id>> extract_tests_from_file(const std::string & fpath_test){
if (fpath_test.empty()){
fprintf(stderr, "%s : No test file found.\n", __func__);
return std::map<std::string, std::vector<gpt_vocab::id>>();