Merge remote-tracking branch 'ggerganov/master'

* ggerganov/master:
  make : fix "main" -> "whisper-cli"
  ci : re-enable Windows cublas build (ggml-org#2676)
  ruby : Fix of C++ header guard name, model URI support, type signature and more (ggml-org#2683)
  examples : handle "main.exe" deprecation

Author: bygreencn

.github/workflows/build.yml

@@ -430,72 +430,73 @@ jobs:
           name: whisper-blas-bin-${{ matrix.arch }}
           path: build/bin/${{ matrix.build }}
 
-# TODO: fix and re-enable
-#  windows-cublas:
-#    runs-on: windows-2019
-#
-#    strategy:
-#      matrix:
-#        build: [Release]
-#        arch: [x64]
-#        cublas: [ON]
-#        sdl2: [ON]
-#        cuda-toolkit: [12.2.0, 11.8.0]
-#        include:
-#          - arch: x64
-#            s2arc: x64
-#          - sdl2: ON
-#            s2ver: 2.28.5
-#
-#    steps:
-#      - name: Clone
-#        uses: actions/checkout@v4
-#
-#      - name: Add msbuild to PATH
-#        uses: microsoft/setup-msbuild@v2
-#
-#      - name: Install CUDA Toolkit
-#        id: cuda-toolkit
-#        uses: Jimver/[email protected]
-#        with:
-#          cuda: '${{ matrix.cuda-toolkit }}'
-#
-#      - name: Fetch SDL2 and set SDL2_DIR
-#        if: matrix.sdl2 == 'ON'
-#        run: |
-#          C:/msys64/usr/bin/wget.exe -qO sdl2.zip https://github.com/libsdl-org/SDL/releases/download/release-${{ matrix.s2ver }}/SDL2-devel-${{ matrix.s2ver }}-VC.zip
-#          7z x sdl2.zip
-#          echo "SDL2_DIR=$env:GITHUB_WORKSPACE/SDL2-${{ matrix.s2ver }}/cmake" >> $env:GITHUB_ENV
-#
-#      - name: Configure
-#        run: >
-#          cmake -S . -B ./build -A ${{ matrix.arch }}
-#          -DCMAKE_BUILD_TYPE=${{ matrix.build }}
-#          -DGGML_CUDA=${{ matrix.cublas }}
-#          -DWHISPER_SDL2=${{ matrix.sdl2 }}
-#
-#      - name: Build ${{ matrix.cuda-toolkit }}
-#        run: |
-#          cd ./build
-#          cmake --build . --config ${{ matrix.build }}
-#
-#      - name: Copy CUDA DLLs
-#        run: >
-#          Copy-Item -PassThru
-#          -Path "${{ steps.cuda-toolkit.outputs.CUDA_PATH }}/bin/*.dll"
-#          -Include cudart64_*,cublas64_*,cublasLt64_*
-#          -Destination build/bin/${{ matrix.build }}
-#
-#      - name: Copy SDL2.dll
-#        if: matrix.sdl2 == 'ON'
-#        run: copy "$env:SDL2_DIR/../lib/${{ matrix.s2arc }}/SDL2.dll" build/bin/${{ matrix.build }}
-#
-#      - name: Upload binaries
-#        if: matrix.sdl2 == 'ON'
-#        uses: actions/upload-artifact@v4
-#        with:
-#          name: whisper-cublas-${{ matrix.cuda-toolkit }}-bin-${{ matrix.arch }}
-#          path: build/bin/${{ matrix.build }}
+  windows-cublas:
+    runs-on: windows-2019
+    strategy:
+      matrix:
+        build: [Release]
+        arch: [x64]
+        cublas: [ON]
+        sdl2: [ON]
+        cuda-toolkit: [12.2.0, 11.8.0]
+        include:
+          - arch: x64
+            sdl2: ON
+            sdl2_ver: 2.28.5
+    steps:
+      - name: Clone repository
+        uses: actions/checkout@v4
+
+      - name: Add msbuild to PATH
+        uses: microsoft/setup-msbuild@v2
+
+      - name: Install CUDA Toolkit
+        id: cuda-toolkit
+        uses: Jimver/[email protected]
+        with:
+          cuda: '${{ matrix.cuda-toolkit }}'
+
+      - name: Install 7-Zip
+        run: choco install 7zip -y
+
+      - name: Fetch SDL2 and set SDL2_DIR
+        if: matrix.sdl2 == 'ON'
+        run: |
+          Invoke-WebRequest -Uri https://github.com/libsdl-org/SDL/releases/download/release-${{ matrix.sdl2_ver }}/SDL2-devel-${{ matrix.sdl2_ver }}-VC.zip -OutFile sdl2.zip
+          7z x sdl2.zip
+          echo "SDL2_DIR=${{ github.workspace }}\SDL2-${{ matrix.sdl2_ver }}\cmake" | Out-File -FilePath $env:GITHUB_ENV -Append
+          echo "${{ github.workspace }}\SDL2-${{ matrix.sdl2_ver }}\cmake" > SDL2_PATH.txt
+ 
+      - name: Configure CMake
+        shell: cmd
+        run: |
+          cmake -S . -B ./build -A ${{ matrix.arch }} ^
+            -DCMAKE_BUILD_TYPE=${{ matrix.build }} ^
+            -DGGML_CUDA=${{ matrix.cublas }} ^
+            -DCMAKE_CUDA_ARCHITECTURES=all  ^
+            -DWHISPER_SDL2=${{ matrix.sdl2 }} ^
+            -DSDL2_DIR="%SDL2_DIR%"
+  
+      - name: Build Project
+        shell: cmd
+        run: |
+          cd ./build
+          cmake --build . --config ${{ matrix.build }}          
+
+      - name: Copy CUDA DLLs
+        run: |
+          Get-ChildItem "${{ steps.cuda-toolkit.outputs.CUDA_PATH }}/bin/" -Filter "*.dll" | 
+          Copy-Item -Destination "build/bin/${{ matrix.build }}"
+
+      - name: Copy SDL2.dll
+        if: matrix.sdl2 == 'ON'
+        run: copy "$env:SDL2_DIR/../lib/${{ matrix.arch }}/SDL2.dll" build/bin/${{ matrix.build }}
+
+      - name: Upload binaries
+        uses: actions/upload-artifact@v4
+        with:
+          name: whisper-cublas-${{ matrix.cuda-toolkit }}-bin-${{ matrix.arch }}
+          path: build/bin/${{ matrix.build }}
 
   emscripten:
     runs-on: ubuntu-latest

Makefile

@@ -64,6 +64,6 @@ tiny.en tiny base.en base small.en small medium.en medium large-v1 large-v2 larg
 		echo "[+] Running $@ on $$f ... (run 'ffplay $$f' to listen)" ; \
 	    echo "----------------------------------------------" ; \
 		echo "" ; \
-		./build/bin/main -m models/ggml-$@.bin -f $$f ; \
+		./build/bin/whisper-cli -m models/ggml-$@.bin -f $$f ; \
 		echo "" ; \
 	done

bindings/ruby/.gitignore

@@ -1,5 +1,3 @@
 LICENSE
 pkg/
-lib/whisper.so
-lib/whisper.bundle
-lib/whisper.dll
+lib/whisper.*

bindings/ruby/README.md

@@ -60,10 +60,10 @@ You also can use shorthand for pre-converted models:
 whisper = Whisper::Context.new("base.en")
 ```
 
-You can see the list of prepared model names by `Whisper::Model.preconverted_models.keys`:
+You can see the list of prepared model names by `Whisper::Model.pre_converted_models.keys`:
 
 ```ruby
-puts Whisper::Model.preconverted_models.keys
+puts Whisper::Model.pre_converted_models.keys
 # tiny
 # tiny.en
 # tiny-q5_1
@@ -87,8 +87,9 @@ whisper = Whisper::Context.new("path/to/your/model.bin")
 Or, you can download model files:
 
 ```ruby
-model_uri = Whisper::Model::URI.new("http://example.net/uri/of/your/model.bin")
-whisper = Whisper::Context.new(model_uri)
+whisper = Whisper::Context.new("https://example.net/uri/of/your/model.bin")
+# Or
+whisper = Whisper::Context.new(URI("https://example.net/uri/of/your/model.bin"))
 ```
 
 See [models][] page for details.
@@ -222,6 +223,17 @@ end
 
 The second argument `samples` may be an array, an object with `length` and `each` method, or a MemoryView. If you can prepare audio data as C array and export it as a MemoryView, whispercpp accepts and works with it with zero copy.
 
+Development
+-----------
+
+    % git clone https://github.com/ggerganov/whisper.cpp.git
+    % cd whisper.cpp/bindings/ruby
+    % rake test
+
+First call of `rake test` builds an extension and downloads a model for testing. After that, you add tests in `tests` directory and modify `ext/ruby_whisper.cpp`.
+
+If something seems wrong on build, running `rake clean` solves some cases.
+
 License
 -------
 

bindings/ruby/ext/ruby_whisper.cpp

@@ -49,6 +49,7 @@ static ID id_length;
 static ID id_next;
 static ID id_new;
 static ID id_to_path;
+static ID id_URI;
 static ID id_pre_converted_models;
 
 static bool is_log_callback_finalized = false;
@@ -283,6 +284,17 @@ static VALUE ruby_whisper_initialize(int argc, VALUE *argv, VALUE self) {
   if (!NIL_P(pre_converted_model)) {
     whisper_model_file_path = pre_converted_model;
   }
+  if (TYPE(whisper_model_file_path) == T_STRING) {
+    const char * whisper_model_file_path_str = StringValueCStr(whisper_model_file_path);
+    if (strncmp("http://", whisper_model_file_path_str, 7) == 0 || strncmp("https://", whisper_model_file_path_str, 8) == 0) {
+      VALUE uri_class = rb_const_get(cModel, id_URI);
+      whisper_model_file_path = rb_class_new_instance(1, &whisper_model_file_path, uri_class);
+    }
+  }
+  if (rb_obj_is_kind_of(whisper_model_file_path, rb_path2class("URI::HTTP"))) {
+    VALUE uri_class = rb_const_get(cModel, id_URI);
+    whisper_model_file_path = rb_class_new_instance(1, &whisper_model_file_path, uri_class);
+  }
   if (rb_respond_to(whisper_model_file_path, id_to_path)) {
     whisper_model_file_path = rb_funcall(whisper_model_file_path, id_to_path, 0);
   }
@@ -837,7 +849,7 @@ static VALUE ruby_whisper_full_get_segment_text(VALUE self, VALUE i_segment) {
 
 /*
  * call-seq:
- *   full_get_segment_no_speech_prob -> Float
+ *   full_get_segment_no_speech_prob(segment_index) -> Float
  */
 static VALUE ruby_whisper_full_get_segment_no_speech_prob(VALUE self, VALUE i_segment) {
   ruby_whisper *rw;
@@ -1755,7 +1767,7 @@ static VALUE ruby_whisper_c_model_type(VALUE self) {
 
 static VALUE ruby_whisper_error_initialize(VALUE self, VALUE code) {
   const int c_code = NUM2INT(code);
-  char *raw_message;
+  const char *raw_message;
   switch (c_code) {
   case -2:
     raw_message = "failed to compute log mel spectrogram";
@@ -1802,6 +1814,7 @@ void Init_whisper() {
   id_next = rb_intern("next");
   id_new = rb_intern("new");
   id_to_path = rb_intern("to_path");
+  id_URI = rb_intern("URI");
   id_pre_converted_models = rb_intern("pre_converted_models");
 
   mWhisper = rb_define_module("Whisper");
@@ -1941,6 +1954,8 @@ void Init_whisper() {
   rb_define_method(cModel, "n_mels", ruby_whisper_c_model_n_mels, 0);
   rb_define_method(cModel, "ftype", ruby_whisper_c_model_ftype, 0);
   rb_define_method(cModel, "type", ruby_whisper_c_model_type, 0);
+
+  rb_require("whisper/model/uri");
 }
 #ifdef __cplusplus
 }

bindings/ruby/ext/ruby_whisper.h

@@ -1,5 +1,5 @@
-#ifndef __RUBY_WHISPER_H
-#define __RUBY_WHISPER_H
+#ifndef RUBY_WHISPER_H
+#define RUBY_WHISPER_H
 
 #include "whisper.h"