Merge tag 'v1.6.2'

* tag 'v1.6.2': release : v1.6.2 Revert "whisper : remove extra backend instance (huh?)" (ggerganov#2182) server : fix typo (ggerganov#2181) ruby : update bindings (ggerganov#2154) release : v1.6.1 examples : add support for decoding input with ffmpeg (Linux) (ggerganov#2133) node : add flash_attn param (ggerganov#2170) ci: Update build.yml to suppress warnings about node.js versions (ggerganov#2166) release : v1.6.0 whisper : use flash attention (ggerganov#2152) talk-llama : reject runs without required arguments (ggerganov#2153) sync : ggml metal : support FA without mask + add asserts (llama/7278) ggml : add RPC backend (llama/6829) rm wait() (llama/7233) CUDA: add FP32 FlashAttention vector kernel (llama/7188) scripts : sync ggml-rpc
bygreencn · Aug 9, 2024 · d579a05 · d579a05
2 parents d545d74 + c7b6988
commit d579a05
Show file tree

Hide file tree

Showing 62 changed files with 16,563 additions and 5,216 deletions.
diff --git a/.github/workflows/build.yml b/.github/workflows/build.yml
@@ -15,10 +15,10 @@ jobs:
 
     steps:
       - name: Clone
-        uses: actions/checkout@v3
+        uses: actions/checkout@v4
 
       - name: Set up QEMU
-        uses: docker/setup-qemu-action@v2
+        uses: docker/setup-qemu-action@v3
 
       - name: Build ${{ matrix.arch }}
         run: |
@@ -36,7 +36,7 @@ jobs:
 
     steps:
       - name: Clone
-        uses: actions/checkout@v3
+        uses: actions/checkout@v4
 
       - name: Dependencies
         run: |
@@ -53,10 +53,10 @@ jobs:
 
     steps:
       - name: Clone
-        uses: actions/checkout@v3
+        uses: actions/checkout@v4
 
       - name: Build
-        uses: cross-platform-actions/action@v0.15.0
+        uses: cross-platform-actions/action@v0.24.0
         with:
           operating_system: freebsd
           version: '13.2'
@@ -77,10 +77,10 @@ jobs:
 
     steps:
       - name: Clone
-        uses: actions/checkout@v3
+        uses: actions/checkout@v4
 
       - name: Set up QEMU
-        uses: docker/setup-qemu-action@v2
+        uses: docker/setup-qemu-action@v3
 
       - name: Build ${{ matrix.arch }}
         run: |
@@ -105,10 +105,10 @@ jobs:
 
     steps:
       - name: Clone
-        uses: actions/checkout@v3
+        uses: actions/checkout@v4
 
       - name: Set up QEMU
-        uses: docker/setup-qemu-action@v2
+        uses: docker/setup-qemu-action@v3
 
       - name: Build ${{ matrix.arch }}
         run: |
@@ -133,10 +133,10 @@ jobs:
 
     steps:
       - name: Clone
-        uses: actions/checkout@v3
+        uses: actions/checkout@v4
 
       - name: Set up QEMU
-        uses: docker/setup-qemu-action@v2
+        uses: docker/setup-qemu-action@v3
 
       - name: Build ${{ matrix.arch }}
         run: |
@@ -165,7 +165,7 @@ jobs:
 
     steps:
       - name: Clone
-        uses: actions/checkout@v3
+        uses: actions/checkout@v4
 
       - name: add oneAPI to apt
         shell: bash
@@ -189,7 +189,7 @@ jobs:
 
       - name: Clone
         id: checkout
-        uses: actions/checkout@v3
+        uses: actions/checkout@v4
 
       - name: Build
         id: cmake_build
@@ -215,7 +215,7 @@ jobs:
 
     steps:
       - name: Clone
-        uses: actions/checkout@v3
+        uses: actions/checkout@v4
 
       - name: add oneAPI to apt
         shell: bash
@@ -239,7 +239,7 @@ jobs:
 
       - name: Clone
         id: checkout
-        uses: actions/checkout@v3
+        uses: actions/checkout@v4
 
       - name: Build
         id: cmake_build
@@ -262,7 +262,7 @@ jobs:
 
     steps:
       - name: Clone
-        uses: actions/checkout@v3
+        uses: actions/checkout@v4
 
       - name: Setup ${{ matrix.sys }}
         uses: msys2/setup-msys2@v2
@@ -328,10 +328,10 @@ jobs:
 
     steps:
       - name: Clone
-        uses: actions/checkout@v3
+        uses: actions/checkout@v4
 
       - name: Add msbuild to PATH
-        uses: microsoft/setup-msbuild@v1
+        uses: microsoft/setup-msbuild@v2
 
       - name: Fetch SDL2 and set SDL2_DIR
         if: matrix.sdl2 == 'ON'
@@ -356,14 +356,14 @@ jobs:
         run: copy "$env:SDL2_DIR/../lib/${{ matrix.s2arc }}/SDL2.dll" build/bin/${{ matrix.build }}
 
       - name: Upload dll
-        uses: actions/upload-artifact@v3
+        uses: actions/upload-artifact@v4
         with:
           name: ${{ matrix.jnaPath }}_whisper.dll
           path: build/bin/${{ matrix.build }}/whisper.dll
 
       - name: Upload binaries
         if: matrix.sdl2 == 'ON'
-        uses: actions/upload-artifact@v1
+        uses: actions/upload-artifact@v4
         with:
           name: whisper-bin-${{ matrix.arch }}
           path: build/bin/${{ matrix.build }}
@@ -392,10 +392,10 @@ jobs:
 
     steps:
       - name: Clone
-        uses: actions/checkout@v3
+        uses: actions/checkout@v4
 
       - name: Add msbuild to PATH
-        uses: microsoft/setup-msbuild@v1
+        uses: microsoft/setup-msbuild@v2
 
       - name: Fetch OpenBLAS
         if: matrix.blas == 'ON'
@@ -453,7 +453,7 @@ jobs:
 
       - name: Upload binaries
         if: matrix.blas == 'ON' && matrix.sdl2 == 'ON'
-        uses: actions/upload-artifact@v1
+        uses: actions/upload-artifact@v4
         with:
           name: whisper-blas${{ matrix.clblast == 'ON' && '-clblast' || ''}}-bin-${{ matrix.arch }}
           path: build/bin/${{ matrix.build }}
@@ -476,14 +476,14 @@ jobs:
 
     steps:
       - name: Clone
-        uses: actions/checkout@v3
+        uses: actions/checkout@v4
 
       - name: Add msbuild to PATH
-        uses: microsoft/setup-msbuild@v1
+        uses: microsoft/setup-msbuild@v2
 
       - name: Install CUDA Toolkit
         id: cuda-toolkit
-        uses: Jimver/[email protected].11
+        uses: Jimver/[email protected].15
         with:
           cuda: '${{ matrix.cuda-toolkit }}'
 
@@ -519,7 +519,7 @@ jobs:
 
       - name: Upload binaries
         if: matrix.sdl2 == 'ON'
-        uses: actions/upload-artifact@v1
+        uses: actions/upload-artifact@v4
         with:
           name: whisper-cublas-${{ matrix.cuda-toolkit }}-bin-${{ matrix.arch }}
           path: build/bin/${{ matrix.build }}
@@ -533,10 +533,10 @@ jobs:
 
     steps:
       - name: Clone
-        uses: actions/checkout@v3
+        uses: actions/checkout@v4
 
       - name: Setup emsdk
-        uses: mymindstorm/setup-emsdk@v12
+        uses: mymindstorm/setup-emsdk@v14
 
       - name: Verify
         run: emcc -v
@@ -555,7 +555,7 @@ jobs:
 
     steps:
       - name: Clone
-        uses: actions/checkout@v3
+        uses: actions/checkout@v4
 
       - name: Configure
         run: |
@@ -573,24 +573,24 @@ jobs:
 
     steps:
       - name: Clone
-        uses: actions/checkout@v3
+        uses: actions/checkout@v4
         with:
           path: whisper
 
       - name: Clone
-        uses: actions/checkout@v3
+        uses: actions/checkout@v4
         with:
           repository: ggerganov/ggml
           path: ggml
 
       - name: Install Java
-        uses: actions/setup-java@v3
+        uses: actions/setup-java@v4
         with:
           distribution: zulu
-          java-version: 17
+          java-version: 21
 
       - name: Setup Android SDK
-        uses: android-actions/setup-android@v2
+        uses: android-actions/setup-android@v3
 
       - name: Build
         run: |
@@ -608,20 +608,19 @@ jobs:
 
     steps:
       - name: Clone
-        uses: actions/checkout@v3
+        uses: actions/checkout@v4
 
       - name: set up JDK 11
-        uses: actions/setup-java@v3
+        uses: actions/setup-java@v4
         with:
           java-version: '11'
           distribution: 'temurin'
           cache: gradle
 
       - name: Setup Android SDK
-        uses: android-actions/setup-android@v2
+        uses: android-actions/setup-android@v3
         with:
-          api-level: 30
-          build-tools-version: 30.0.3
+          cmdline-tools-version: 9.0
 
       - name: Build
         run: |
@@ -633,15 +632,16 @@ jobs:
     needs: [ 'windows' ]
     runs-on: windows-latest
     steps:
-      - uses: actions/checkout@v3
+      - uses: actions/checkout@v4
 
       - name: Install Java
-        uses: actions/setup-java@v1
+        uses: actions/setup-java@v4
         with:
-          java-version: 17
+          distribution: zulu
+          java-version: 20
 
       - name: Download Windows lib
-        uses: actions/download-artifact@v3
+        uses: actions/download-artifact@v4
         with:
           name: win32-x86-64_whisper.dll
           path: bindings/java/build/generated/resources/main/win32-x86-64
@@ -654,7 +654,7 @@ jobs:
           ./gradlew build
 
       - name: Upload jar
-        uses: actions/upload-artifact@v3
+        uses: actions/upload-artifact@v4
         with:
           name: whispercpp.jar
           path: bindings/java/build/libs/whispercpp-*.jar
@@ -676,7 +676,7 @@ jobs:
 
     steps:
       - name: Clone
-        uses: actions/checkout@v3
+        uses: actions/checkout@v4
 
       - name: Test quantize
         run: |

diff --git a/CMakeLists.txt b/CMakeLists.txt
@@ -3,7 +3,7 @@ cmake_minimum_required (VERSION 3.5)
 # Allow for the creation of solution folders.
 set_property(GLOBAL PROPERTY USE_FOLDERS ON)
 
-project(whisper.cpp VERSION 1.5.5)
+project(whisper.cpp VERSION 1.6.2)
 set(SOVERSION 1)
 
 # Add path to modules
@@ -64,6 +64,10 @@ option(WHISPER_SampleRate             "whisper: support for libSampleRate" OFF)
 option(WHISPER_sndfile                "whisper: support for libsndfile" OFF)
 
 
+if (CMAKE_SYSTEM_NAME MATCHES "Linux")
+    option(WHISPER_FFMPEG                 "whisper: support building and linking with ffmpeg libs (avcodec, swresample, ...)" OFF)
+endif()
+
 option(WHISPER_NO_AVX                 "whisper: disable AVX"         OFF)
 option(WHISPER_NO_AVX2                "whisper: disable AVX2"        OFF)
 option(WHISPER_NO_AVX512              "whisper: disable AVX512"      ON)
@@ -130,6 +134,26 @@ else()
     set(CMAKE_CXX_STANDARD 11)
 endif()
 
+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)
+    set(WHISPER_EXTRA_LIBS  ${WHISPER_EXTRA_LIBS} ${FFMPEG_LIBRARIES})
+endif()
+
 # on APPLE
 if (APPLE)
     # include Accelerate framework

diff --git a/README.md b/README.md
@@ -6,7 +6,7 @@
 [![License: MIT](https://img.shields.io/badge/license-MIT-blue.svg)](https://opensource.org/licenses/MIT)
 [![npm](https://img.shields.io/npm/v/whisper.cpp.svg)](https://www.npmjs.com/package/whisper.cpp/)
 
-Stable: [v1.5.5](https://github.com/ggerganov/whisper.cpp/releases/tag/v1.5.5) / [Roadmap | F.A.Q.](https://github.com/ggerganov/whisper.cpp/discussions/126)
+Stable: [v1.6.2](https://github.com/ggerganov/whisper.cpp/releases/tag/v1.6.0) / [Roadmap | F.A.Q.](https://github.com/ggerganov/whisper.cpp/discussions/126)
 
 High-performance inference of [OpenAI's Whisper](https://github.com/openai/whisper) automatic speech recognition (ASR) model:
 

diff --git a/bindings/ios b/bindings/ios
diff --git a/bindings/javascript/package.json b/bindings/javascript/package.json
@@ -1,6 +1,6 @@
 {
   "name": "whisper.cpp",
-  "version": "1.5.5",
+  "version": "1.6.2",
   "description": "Whisper speech recognition",
   "main": "whisper.js",
   "scripts": {

diff --git a/bindings/ruby/Rakefile b/bindings/ruby/Rakefile
@@ -0,0 +1,12 @@
+require 'rake/clean'
+  require 'rubygems/package'
+
+desc 'Build gem'
+task :package do
+  spec_source = File.read File.join(File.dirname(__FILE__),'whispercpp.gemspec')
+  spec = nil
+  # see: http://gist.github.com/16215
+  Thread.new { spec = eval("#{spec_source}") }.join
+  spec.validate
+  Gem::Package.build(spec)
+end