Add WebVTT-in-video-stream support (#196)

* Fix `whisper_buffer` and `resampled_buffer` data race

`media_unpause` was causing `wisper_buffer` to be freed while
`vad_based_segmentation`/`hybrid_vad_segmentation` need that buffer
to not be modified for the duration of those calls

* Slightly improve handling for weird subtitle output filenames

* Squashed 'deps/c-webvtt-in-video-stream/' content from commit 5579ca6

git-subtree-dir: deps/c-webvtt-in-video-stream
git-subtree-split: 5579ca6dc9dcf94e3c14631c6c01b2ee4dfcf005

* Add WIP webvtt sei functionality

* Add webvtt recording/streaming settings

* Make latency_to_video_in_msecs and send_frequency_hz configurable

* Make webvtt languages configurable

* Add translation and main language separately

* Add rust CI integration

.github/workflows/build-project.yaml

@@ -119,6 +119,16 @@ jobs:
           restore-keys: |
             ${{ runner.os }}-ccache-${{ matrix.architecture }}-
 
+      - uses: actions-rust-lang/setup-rust-toolchain@v1
+        if: matrix.architecture == 'arm64'
+        with:
+          target: aarch64-apple-darwin
+
+      - uses: actions-rust-lang/setup-rust-toolchain@v1
+        if: matrix.architecture == 'x86_64'
+        with:
+          target: x86_64-apple-darwin
+
       - name: Set Up Codesigning 🔑
         uses: ./.github/actions/setup-macos-codesigning
         if: fromJSON(needs.check-event.outputs.codesign)
@@ -197,6 +207,8 @@ jobs:
           echo "pluginName=${product_name}" >> $GITHUB_OUTPUT
           echo "pluginVersion=${product_version}" >> $GITHUB_OUTPUT
 
+      - uses: actions-rust-lang/setup-rust-toolchain@v1
+
       - uses: actions/cache@v4
         id: ccache-cache
         with:
@@ -271,6 +283,8 @@ jobs:
           "pluginName=${ProductName}" >> $env:GITHUB_OUTPUT
           "pluginVersion=${ProductVersion}" >> $env:GITHUB_OUTPUT
 
+      - uses: actions-rust-lang/setup-rust-toolchain@v1
+
       - name: Build Plugin 🧱
         uses: ./.github/actions/build-plugin
         with:

.github/workflows/check-format.yaml

@@ -25,3 +25,17 @@ jobs:
         uses: ./.github/actions/run-cmake-format
         with:
           failCondition: error
+
+  cargo-fmt:
+    runs-on: ubuntu-22.04
+    steps:
+      - uses: actions/checkout@v4
+        with:
+          fetch-depth: 0
+      - uses: actions-rust-lang/setup-rust-toolchain@v1
+        with:
+          components: rustfmt
+      - name: rustfmt
+        uses: actions-rust-lang/rustfmt@v1
+        with:
+          manifest-path: deps/c-webvtt-in-video-stream/Cargo.toml

CMakeLists.txt

@@ -106,6 +106,14 @@ if(DEFINED ENV{LOCALVOCAL_EXTRA_VERBOSE})
   target_compile_definitions(${CMAKE_PROJECT_NAME} PRIVATE LOCALVOCAL_EXTRA_VERBOSE)
 endif()
 
+option(ENABLE_WEBVTT "Enable WebVTT embedding" ON)
+
+if(ENABLE_WEBVTT)
+  include(cmake/BuildWebVTT.cmake)
+  target_link_libraries(${CMAKE_PROJECT_NAME} PRIVATE c_webvtt_in_video_stream)
+  target_compile_definitions(c_webvtt_in_video_stream INTERFACE ENABLE_WEBVTT)
+endif()
+
 target_sources(
   ${CMAKE_PROJECT_NAME}
   PRIVATE src/plugin-main.c

cmake/BuildWebVTT.cmake

@@ -0,0 +1,23 @@
+include(FetchContent)
+
+set(Rust_RUSTUP_INSTALL_MISSING_TARGET true)
+
+if(OS_MACOS)
+  if("$ENV{MACOS_ARCH}" STREQUAL "x86_64")
+    set(Rust_CARGO_TARGET "x86_64-apple-darwin")
+  endif()
+endif()
+
+FetchContent_Declare(
+  Corrosion
+  GIT_REPOSITORY https://github.com/corrosion-rs/corrosion.git
+  GIT_TAG v0.5 # Optionally specify a commit hash, version tag or branch here
+)
+FetchContent_MakeAvailable(Corrosion)
+
+# Import targets defined in a package or workspace manifest `Cargo.toml` file
+corrosion_import_crate(MANIFEST_PATH "${CMAKE_SOURCE_DIR}/deps/c-webvtt-in-video-stream/Cargo.toml" CRATE_TYPES
+                       "staticlib" PROFILE release)
+
+set_target_properties(c_webvtt_in_video_stream PROPERTIES INTERFACE_INCLUDE_DIRECTORIES
+                                                          "${CMAKE_SOURCE_DIR}/deps/c-webvtt-in-video-stream/target/")

data/locale/en-US.ini

@@ -4,6 +4,12 @@ vad_threshold="VAD Threshold"
 log_level="Internal Log Level"
 log_words="Log Output to Console"
 caption_to_stream="Stream Captions"
+webvtt_group="WebVTT"
+webvtt_caption_to_stream="Add WebVTT captions to stream"
+webvtt_caption_to_recording="Add WebVTT captions to recording"
+webvtt_latency_to_video_in_msecs="Latency to video (milliseconds)"
+webvtt_send_frequency_hz="Send frequency (Hz)"
+webvtt_language_description="Language $1"
 subtitle_sources="Output Destination"
 none_no_output="None / No output"
 file_output_enable="Save to File"

deps/c-webvtt-in-video-stream/.gitignore

@@ -0,0 +1 @@
+/target

deps/c-webvtt-in-video-stream/.vscode/settings.json

@@ -0,0 +1,7 @@
+{
+    "editor.formatOnSave": true,
+    "evenBetterToml.formatter.reorderKeys": true,
+    "evenBetterToml.formatter.reorderArrays": true,
+    "evenBetterToml.formatter.trailingNewline": true,
+    "rust-analyzer.check.command": "clippy"
+}