whispercpp 1.3.1 → 1.3.2

data/ext/sources/CMakeLists.txt

@@ -0,0 +1,251 @@
+cmake_minimum_required(VERSION 3.5) # for add_link_options and implicit target directories.
+project("whisper.cpp" C CXX)
+project("whisper.cpp" VERSION 1.7.5)
+include(CheckIncludeFileCXX)
+
+set(SOVERSION 1)
+
+#set(CMAKE_WARN_DEPRECATED YES)
+set(CMAKE_WARN_UNUSED_CLI YES)
+
+set(CMAKE_EXPORT_COMPILE_COMMANDS ON)
+
+if (NOT XCODE AND NOT MSVC AND NOT CMAKE_BUILD_TYPE)
+    set(CMAKE_BUILD_TYPE Release CACHE STRING "Build type" FORCE)
+    set_property(CACHE CMAKE_BUILD_TYPE PROPERTY STRINGS "Debug" "Release" "MinSizeRel" "RelWithDebInfo")
+endif()
+
+# Add path to modules
+list(APPEND CMAKE_MODULE_PATH "${CMAKE_CURRENT_SOURCE_DIR}/cmake/")
+
+set(CMAKE_RUNTIME_OUTPUT_DIRECTORY ${CMAKE_BINARY_DIR}/bin)
+
+if (CMAKE_SOURCE_DIR STREQUAL CMAKE_CURRENT_SOURCE_DIR)
+    set(WHISPER_STANDALONE ON)
+
+    include(git-vars)
+
+    # configure project version
+    configure_file(${CMAKE_SOURCE_DIR}/bindings/javascript/package-tmpl.json ${CMAKE_SOURCE_DIR}/bindings/javascript/package.json @ONLY)
+else()
+    set(WHISPER_STANDALONE OFF)
+endif()
+
+if (EMSCRIPTEN)
+    set(BUILD_SHARED_LIBS_DEFAULT OFF)
+
+    option(WHISPER_WASM_SINGLE_FILE "whisper: embed WASM inside the generated whisper.js" ON)
+
+    # TODO: without these, we get the following error:
+    #       wasm-ld: error: --shared-memory is disallowed by whisper.cpp.o because it was not compiled with 'atomics' or 'bulk-memory' features.
+    set(CMAKE_C_FLAGS   "${CMAKE_C_FLAGS}   -pthread")
+    set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -pthread")
+
+    set(CMAKE_EXE_LINKER_FLAGS "${CMAKE_EXE_LINKER_FLAGS} -s TOTAL_STACK=5242880")
+    set(CMAKE_SHARED_LINKER_FLAGS "${CMAKE_SHARED_LINKER_FLAGS} -s TOTAL_STACK=5242880")
+
+    set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -Wno-deprecated")
+else()
+    if (MINGW)
+        set(BUILD_SHARED_LIBS_DEFAULT OFF)
+    else()
+        set(BUILD_SHARED_LIBS_DEFAULT ON)
+    endif()
+endif()
+
+option(BUILD_SHARED_LIBS "build shared libraries" ${BUILD_SHARED_LIBS_DEFAULT})
+
+#
+# option list
+#
+
+# debug
+option(WHISPER_ALL_WARNINGS           "whisper: enable all compiler warnings"                   ON)
+option(WHISPER_ALL_WARNINGS_3RD_PARTY "whisper: enable all compiler warnings in 3rd party libs" OFF)
+
+# build
+option(WHISPER_FATAL_WARNINGS  "whisper: enable -Werror flag"               OFF)
+option(WHISPER_USE_SYSTEM_GGML "whisper: use system-installed GGML library" OFF)
+
+# sanitizers
+option(WHISPER_SANITIZE_THREAD    "whisper: enable thread sanitizer"    OFF)
+option(WHISPER_SANITIZE_ADDRESS   "whisper: enable address sanitizer"   OFF)
+option(WHISPER_SANITIZE_UNDEFINED "whisper: enable undefined sanitizer" OFF)
+
+# extra artifacts
+option(WHISPER_BUILD_TESTS    "whisper: build tests"          ${WHISPER_STANDALONE})
+option(WHISPER_BUILD_EXAMPLES "whisper: build examples"       ${WHISPER_STANDALONE})
+option(WHISPER_BUILD_SERVER   "whisper: build server example" ${WHISPER_STANDALONE})
+
+# 3rd party libs
+option(WHISPER_CURL "whisper: use libcurl to download model from an URL" OFF)
+option(WHISPER_SDL2 "whisper: support for libSDL2" OFF)
+
+if (CMAKE_SYSTEM_NAME MATCHES "Linux")
+    option(WHISPER_FFMPEG "whisper: support building and linking with ffmpeg libs (avcodec, swresample, ...)" OFF)
+endif()
+
+option(WHISPER_COREML                "whisper: enable Core ML framework"  OFF)
+option(WHISPER_COREML_ALLOW_FALLBACK "whisper: allow non-CoreML fallback" OFF)
+option(WHISPER_OPENVINO              "whisper: support for OpenVINO"      OFF)
+
+# Required for relocatable CMake package
+include(${CMAKE_CURRENT_SOURCE_DIR}/cmake/build-info.cmake)
+
+# override ggml options
+set(GGML_SANITIZE_THREAD    ${WHISPER_SANITIZE_THREAD})
+set(GGML_SANITIZE_ADDRESS   ${WHISPER_SANITIZE_ADDRESS})
+set(GGML_SANITIZE_UNDEFINED ${WHISPER_SANITIZE_UNDEFINED})
+set(GGML_ALL_WARNINGS       ${WHISPER_ALL_WARNINGS})
+set(GGML_FATAL_WARNINGS     ${WHISPER_FATAL_WARNINGS})
+
+# transition helpers
+function (whisper_option_depr TYPE OLD NEW)
+    if (${OLD})
+        message(${TYPE} "${OLD} is deprecated and will be removed in the future.\nUse ${NEW} instead\n")
+        set(${NEW} ON)
+    endif()
+endfunction()
+
+whisper_option_depr(FATAL_ERROR WHISPER_CUBLAS              GGML_CUDA)
+whisper_option_depr(WARNING     WHISPER_CUDA                GGML_CUDA)
+whisper_option_depr(WARNING     WHISPER_KOMPUTE             GGML_KOMPUTE)
+whisper_option_depr(WARNING     WHISPER_METAL               GGML_METAL)
+whisper_option_depr(WARNING     WHISPER_METAL_EMBED_LIBRARY GGML_METAL_EMBED_LIBRARY)
+whisper_option_depr(WARNING     WHISPER_NATIVE              GGML_NATIVE)
+whisper_option_depr(WARNING     WHISPER_OPENMP              GGML_OPENMP)
+whisper_option_depr(WARNING     WHISPER_RPC                 GGML_RPC)
+whisper_option_depr(WARNING     WHISPER_SYCL                GGML_SYCL)
+whisper_option_depr(WARNING     WHISPER_SYCL_F16            GGML_SYCL_F16)
+whisper_option_depr(WARNING     WHISPER_CCACHE              GGML_CCACHE)
+
+if (GGML_CUDA AND NOT MSVC)
+    #GGML_CUDA enabled, add the necessary compile options -Wno-deprecated-gpu-targets
+    add_compile_options(-Wno-deprecated-gpu-targets)
+endif()
+
+#
+# build the library
+#
+
+if (NOT TARGET ggml)
+    if (WHISPER_USE_SYSTEM_GGML)
+        find_package(ggml REQUIRED)
+        if (NOT ggml_FOUND)
+            message(FATAL_ERROR "System-installed GGML library not found.")
+        endif()
+        add_library(ggml ALIAS ggml::ggml)
+    else()
+        add_subdirectory(ggml)
+        if(WIN32)
+            # The following adds a _DISABLE_CONSTEXPR_MUTEX_CONSTRUCTOR macro and is a workaround for
+            # the Windows C++ standard library which does not support constexpr mutexes.
+            # From the release notes://github.com/microsoft/STL/wiki/Changelog
+            #  Disable constexpr mutex constructor on Windows
+            #  Fixed mutex's constructor to be constexpr. #3824 #4000 #4339
+            #  Note: Programs that aren't following the documented restrictions on binary compatibility may encounter
+            #  null dereferences in mutex machinery. You must follow this rule:
+            #  When you mix binaries built by different supported versions of the toolset, the Redistributable version
+            #  must be at least as new as the latest toolset used by any app component.
+            #  You can define _DISABLE_CONSTEXPR_MUTEX_CONSTRUCTOR as an escape hatch.
+            #
+            # Specifically to whisper.cpp this would cause a crash when using the Java bindings.
+            # resulting in a Invalid memory access error.
+            target_compile_definitions(ggml-base PRIVATE _DISABLE_CONSTEXPR_MUTEX_CONSTRUCTOR)
+        endif()
+    endif()
+    # ... otherwise assume ggml is added by a parent CMakeLists.txt
+endif()
+add_subdirectory(src)
+
+#
+# install
+#
+
+include(GNUInstallDirs)
+include(CMakePackageConfigHelpers)
+
+set(WHISPER_BUILD_NUMBER        ${BUILD_NUMBER})
+set(WHISPER_BUILD_COMMIT        ${BUILD_COMMIT})
+set(WHISPER_INSTALL_VERSION     ${CMAKE_PROJECT_VERSION})
+
+set(WHISPER_INCLUDE_INSTALL_DIR ${CMAKE_INSTALL_INCLUDEDIR} CACHE PATH "Location of header  files")
+set(WHISPER_LIB_INSTALL_DIR     ${CMAKE_INSTALL_LIBDIR}     CACHE PATH "Location of library files")
+set(WHISPER_BIN_INSTALL_DIR     ${CMAKE_INSTALL_BINDIR}     CACHE PATH "Location of binary  files")
+
+get_directory_property(WHISPER_TRANSIENT_DEFINES COMPILE_DEFINITIONS)
+
+set_target_properties(whisper PROPERTIES PUBLIC_HEADER ${CMAKE_CURRENT_SOURCE_DIR}/include/whisper.h)
+install(TARGETS whisper LIBRARY PUBLIC_HEADER)
+
+configure_package_config_file(
+        ${CMAKE_CURRENT_SOURCE_DIR}/cmake/whisper-config.cmake.in
+        ${CMAKE_CURRENT_BINARY_DIR}/whisper-config.cmake
+    INSTALL_DESTINATION ${CMAKE_INSTALL_LIBDIR}/cmake/whisper
+    PATH_VARS
+    WHISPER_INCLUDE_INSTALL_DIR
+    WHISPER_LIB_INSTALL_DIR
+    WHISPER_BIN_INSTALL_DIR )
+
+write_basic_package_version_file(
+    ${CMAKE_CURRENT_BINARY_DIR}/whisper-version.cmake
+    VERSION ${WHISPER_INSTALL_VERSION}
+    COMPATIBILITY SameMajorVersion)
+
+install(FILES ${CMAKE_CURRENT_BINARY_DIR}/whisper-config.cmake
+              ${CMAKE_CURRENT_BINARY_DIR}/whisper-version.cmake
+        DESTINATION ${CMAKE_INSTALL_LIBDIR}/cmake/whisper)
+
+configure_file(cmake/whisper.pc.in
+        "${CMAKE_CURRENT_BINARY_DIR}/whisper.pc"
+        @ONLY)
+
+install(FILES "${CMAKE_CURRENT_BINARY_DIR}/whisper.pc"
+        DESTINATION lib/pkgconfig)
+
+#
+# programs, examples and tests
+#
+
+if (WHISPER_BUILD_TESTS AND NOT CMAKE_JS_VERSION)
+    include(CTest)
+    add_subdirectory(tests)
+endif ()
+
+if (WHISPER_BUILD_EXAMPLES)
+    add_subdirectory(examples)
+endif()
+
+if (MSVC)
+    set(MSVC_WARNING_FLAGS
+        /wd4101  # Unreferenced local variable
+        /wd4005  # Macro redefinition
+        /wd4065  # switch statement contains 'default' but no 'case' labels
+        /wd4267  # Conversion from 'size_t' to a smaller type, possible loss of data
+        /wd4244  # Conversion from one type to another type, possible loss of ata
+        /wd4805  # Unsafe mix of type
+        /wd4305  # Truncation from 'type1' to 'type2' (often double to float)
+        /wd4996  # Function or variable may be unsafe/deprecated
+    )
+    function(disable_msvc_warnings target_name)
+        if(TARGET ${target_name})
+            target_compile_options(${target_name} PRIVATE ${MSVC_WARNING_FLAGS})
+        endif()
+    endfunction()
+
+    if (WHISPER_BUILD_EXAMPLES)
+        disable_msvc_warnings(whisper)
+        disable_msvc_warnings(common)
+        disable_msvc_warnings(common-sdl)
+        disable_msvc_warnings(lsp)
+        disable_msvc_warnings(wchess-core)
+        disable_msvc_warnings(whisper-command)
+        disable_msvc_warnings(whisper-cli)
+        disable_msvc_warnings(whisper-server)
+        disable_msvc_warnings(whisper-stream)
+        disable_msvc_warnings(whisper-talk-llama)
+        disable_msvc_warnings(whisper-bench)
+        disable_msvc_warnings(quantize)
+        disable_msvc_warnings(vad-speech-segments)
+    endif()
+endif()

data/ext/sources/bindings/javascript/CMakeLists.txt

@@ -0,0 +1,41 @@
+set(TARGET libwhisper)
+
+add_executable(${TARGET}
+    emscripten.cpp
+    )
+
+target_link_libraries(${TARGET} PRIVATE
+    whisper
+    )
+
+unset(EXTRA_FLAGS)
+
+if (WHISPER_WASM_SINGLE_FILE)
+    set(EXTRA_FLAGS "-s SINGLE_FILE=1")
+    message(STATUS "Embedding WASM inside whisper.js")
+
+    add_custom_command(
+        TARGET ${TARGET} POST_BUILD
+        COMMAND ${CMAKE_COMMAND} -E copy
+        ${CMAKE_BINARY_DIR}/bin/libwhisper.js
+        ${CMAKE_CURRENT_SOURCE_DIR}/whisper.js
+        )
+
+    add_custom_command(
+        TARGET ${TARGET} POST_BUILD
+        COMMAND ${CMAKE_COMMAND} -E copy
+        ${CMAKE_BINARY_DIR}/bin/libwhisper.worker.js
+        ${CMAKE_CURRENT_SOURCE_DIR}/libwhisper.worker.js
+        )
+endif()
+
+set_target_properties(${TARGET} PROPERTIES LINK_FLAGS " \
+    --bind \
+    -s MODULARIZE=1 \
+    -s EXPORT_NAME=\"'whisper_factory'\" \
+    -s FORCE_FILESYSTEM=1 \
+    -s USE_PTHREADS=1 \
+    -s PTHREAD_POOL_SIZE=8 \
+    -s ALLOW_MEMORY_GROWTH=1 \
+    ${EXTRA_FLAGS} \
+    ")

data/ext/sources/bindings/javascript/emscripten.cpp

@@ -0,0 +1,93 @@
+//
+// This is the Javascript API of whisper.cpp
+//
+// Very crude at the moment.
+// Feel free to contribute and make this better!
+//
+// See the tests/test-whisper.js for sample usage
+//
+
+#include "whisper.h"
+
+#include <emscripten.h>
+#include <emscripten/bind.h>
+
+#include <thread>
+#include <vector>
+
+struct whisper_context * g_context;
+
+EMSCRIPTEN_BINDINGS(whisper) {
+    emscripten::function("init", emscripten::optional_override([](const std::string & path_model) {
+        if (g_context == nullptr) {
+            g_context = whisper_init_from_file_with_params(path_model.c_str(), whisper_context_default_params());
+            if (g_context != nullptr) {
+                return true;
+            } else {
+                return false;
+            }
+        }
+
+        return false;
+    }));
+
+    emscripten::function("free", emscripten::optional_override([]() {
+        if (g_context) {
+            whisper_free(g_context);
+            g_context = nullptr;
+        }
+    }));
+
+    emscripten::function("full_default", emscripten::optional_override([](const emscripten::val & audio, const std::string & lang, bool translate) {
+        if (g_context == nullptr) {
+            return -1;
+        }
+
+        struct whisper_full_params params = whisper_full_default_params(whisper_sampling_strategy::WHISPER_SAMPLING_GREEDY);
+
+        params.print_realtime   = true;
+        params.print_progress   = false;
+        params.print_timestamps = true;
+        params.print_special    = false;
+        params.translate        = translate;
+        params.language         = whisper_is_multilingual(g_context) ? lang.c_str() : "en";
+        params.n_threads        = std::min(8, (int) std::thread::hardware_concurrency());
+        params.offset_ms        = 0;
+
+        std::vector<float> pcmf32;
+        const int n = audio["length"].as<int>();
+
+        emscripten::val heap = emscripten::val::module_property("HEAPU8");
+        emscripten::val memory = heap["buffer"];
+
+        pcmf32.resize(n);
+
+        emscripten::val memoryView = audio["constructor"].new_(memory, reinterpret_cast<uintptr_t>(pcmf32.data()), n);
+        memoryView.call<void>("set", audio);
+
+        // print system information
+        {
+            printf("\n");
+            printf("system_info: n_threads = %d / %d | %s\n",
+                    params.n_threads, std::thread::hardware_concurrency(), whisper_print_system_info());
+
+            printf("\n");
+            printf("%s: processing %d samples, %.1f sec, %d threads, %d processors, lang = %s, task = %s ...\n",
+                    __func__, int(pcmf32.size()), float(pcmf32.size())/WHISPER_SAMPLE_RATE,
+                    params.n_threads, 1,
+                    params.language,
+                    params.translate ? "translate" : "transcribe");
+
+            printf("\n");
+        }
+
+        // run whisper
+        {
+            whisper_reset_timings(g_context);
+            whisper_full(g_context, params, pcmf32.data(), pcmf32.size());
+            whisper_print_timings(g_context);
+        }
+
+        return 0;
+    }));
+}

data/ext/sources/bindings/javascript/libwhisper.worker.js

@@ -0,0 +1 @@
+"use strict";var Module={};var ENVIRONMENT_IS_NODE=typeof process=="object"&&typeof process.versions=="object"&&typeof process.versions.node=="string";if(ENVIRONMENT_IS_NODE){var nodeWorkerThreads=require("worker_threads");var parentPort=nodeWorkerThreads.parentPort;parentPort.on("message",data=>onmessage({data:data}));var fs=require("fs");Object.assign(global,{self:global,require:require,Module:Module,location:{href:__filename},Worker:nodeWorkerThreads.Worker,importScripts:f=>(0,eval)(fs.readFileSync(f,"utf8")+"//# sourceURL="+f),postMessage:msg=>parentPort.postMessage(msg),performance:global.performance||{now:Date.now}})}var initializedJS=false;function threadPrintErr(){var text=Array.prototype.slice.call(arguments).join(" ");if(ENVIRONMENT_IS_NODE){fs.writeSync(2,text+"\n");return}console.error(text)}function threadAlert(){var text=Array.prototype.slice.call(arguments).join(" ");postMessage({cmd:"alert",text:text,threadId:Module["_pthread_self"]()})}var err=threadPrintErr;self.alert=threadAlert;Module["instantiateWasm"]=(info,receiveInstance)=>{var module=Module["wasmModule"];Module["wasmModule"]=null;var instance=new WebAssembly.Instance(module,info);return receiveInstance(instance)};self.onunhandledrejection=e=>{throw e.reason||e};function handleMessage(e){try{if(e.data.cmd==="load"){let messageQueue=[];self.onmessage=e=>messageQueue.push(e);self.startWorker=instance=>{Module=instance;postMessage({"cmd":"loaded"});for(let msg of messageQueue){handleMessage(msg)}self.onmessage=handleMessage};Module["wasmModule"]=e.data.wasmModule;for(const handler of e.data.handlers){Module[handler]=(...args)=>{postMessage({cmd:"callHandler",handler:handler,args:args})}}Module["wasmMemory"]=e.data.wasmMemory;Module["buffer"]=Module["wasmMemory"].buffer;Module["ENVIRONMENT_IS_PTHREAD"]=true;if(typeof e.data.urlOrBlob=="string"){importScripts(e.data.urlOrBlob)}else{var objectUrl=URL.createObjectURL(e.data.urlOrBlob);importScripts(objectUrl);URL.revokeObjectURL(objectUrl)}whisper_factory(Module)}else if(e.data.cmd==="run"){Module["__emscripten_thread_init"](e.data.pthread_ptr,0,0,1);Module["__emscripten_thread_mailbox_await"](e.data.pthread_ptr);Module["establishStackSpace"]();Module["PThread"].receiveObjectTransfer(e.data);Module["PThread"].threadInitTLS();if(!initializedJS){Module["__embind_initialize_bindings"]();initializedJS=true}try{Module["invokeEntryPoint"](e.data.start_routine,e.data.arg)}catch(ex){if(ex!="unwind"){throw ex}}}else if(e.data.cmd==="cancel"){if(Module["_pthread_self"]()){Module["__emscripten_thread_exit"](-1)}}else if(e.data.target==="setimmediate"){}else if(e.data.cmd==="checkMailbox"){if(initializedJS){Module["checkMailbox"]()}}else if(e.data.cmd){err(`worker.js received unknown command ${e.data.cmd}`);err(e.data)}}catch(ex){if(Module["__emscripten_thread_crashed"]){Module["__emscripten_thread_crashed"]()}throw ex}}self.onmessage=handleMessage;

data/ext/sources/bindings/javascript/package-tmpl.json

@@ -0,0 +1,26 @@
+{
+  "name": "whisper.cpp",
+  "version": "@PROJECT_VERSION@",
+  "description": "Whisper speech recognition",
+  "main": "whisper.js",
+  "scripts": {
+    "test": "echo \"todo: add tests\" && exit 0"
+  },
+  "repository": {
+    "type": "git",
+    "url": "git+https://github.com/ggerganov/whisper.cpp"
+  },
+  "keywords": [
+    "openai",
+    "whisper",
+    "speech-to-text",
+    "speech-recognition",
+    "transformer"
+  ],
+  "author": "Georgi Gerganov",
+  "license": "MIT",
+  "bugs": {
+    "url": "https://github.com/ggerganov/whisper.cpp/issues"
+  },
+  "homepage": "https://github.com/ggerganov/whisper.cpp#readme"
+}

data/ext/sources/bindings/javascript/package.json

@@ -0,0 +1,26 @@
+{
+  "name": "whisper.cpp",
+  "version": "1.7.5",
+  "description": "Whisper speech recognition",
+  "main": "whisper.js",
+  "scripts": {
+    "test": "echo \"todo: add tests\" && exit 0"
+  },
+  "repository": {
+    "type": "git",
+    "url": "git+https://github.com/ggerganov/whisper.cpp"
+  },
+  "keywords": [
+    "openai",
+    "whisper",
+    "speech-to-text",
+    "speech-recognition",
+    "transformer"
+  ],
+  "author": "Georgi Gerganov",
+  "license": "MIT",
+  "bugs": {
+    "url": "https://github.com/ggerganov/whisper.cpp/issues"
+  },
+  "homepage": "https://github.com/ggerganov/whisper.cpp#readme"
+}