tests : update VAD tests to use Silero V6.2.0 (#3534)

* tests : update VAD tests to use Silero V6.2.0

This commit updates the VAD tests to use the Silero V6.2.0 instead of
V5.1.2. I'm was not sure if we needed to keep testing for both versions,
but opted to just update to the latest version for simplicity.

* wasm : use C++17 for emscripten builds

This commit updates the CMakeLists.txt file to explicitly set the C++
standard to C++17 when building with Emscripten.

The motivation for this change is that building with Emscripten
will currently fail locally and on CI with the following error:
```console
[ 75%] Building CXX object examples/CMakeFiles/common.dir/common-ggml.cpp.o
In file included from /home/danbev/work/ai/whisper.cpp/examples/stream.wasm/emscripten.cpp:5:
/home/danbev/work/utils/emsdk/upstream/emscripten/cache/sysroot/include/emscripten/bind.h:11:2: error:
      "embind requires -std=c++17 or newer"
   11 | #error "embind requires -std=c++17 or newer"
      |  ^
In file included from /home/danbev/work/ai/whisper.cpp/examples/whisper.wasm/emscripten.cpp:4:
/home/danbev/work/utils/emsdk/upstream/emscripten/cache/sysroot/include/emscripten/bind.h:11:2: error:
      "embind requires -std=c++17 or newer"
   11 | #error "embind requires -std=c++17 or newer"
      |  ^
```

CMakeLists.txt

@@ -34,6 +34,9 @@ endif()
 if (EMSCRIPTEN)
     set(BUILD_SHARED_LIBS_DEFAULT OFF)
 
+    set(CMAKE_CXX_STANDARD 17)
+    set(CMAKE_CXX_STANDARD_REQUIRED ON)
+
     option(WHISPER_WASM_SINGLE_FILE "whisper: embed WASM inside the generated whisper.js" ON)
 
     # TODO: without these, we get the following error:

tests/CMakeLists.txt

@@ -94,7 +94,7 @@ add_executable(${VAD_TEST} ${VAD_TEST}.cpp)
 target_include_directories(${VAD_TEST} PRIVATE ../include ../ggml/include ../examples)
 target_link_libraries(${VAD_TEST} PRIVATE common)
 target_compile_definitions(${VAD_TEST} PRIVATE
-    VAD_MODEL_PATH="${PROJECT_SOURCE_DIR}/models/for-tests-silero-v5.1.2-ggml.bin"
+    VAD_MODEL_PATH="${PROJECT_SOURCE_DIR}/models/for-tests-silero-v6.2.0-ggml.bin"
     SAMPLE_PATH="${PROJECT_SOURCE_DIR}/samples/jfk.wav")
 add_test(NAME ${VAD_TEST} COMMAND ${VAD_TEST})
 set_tests_properties(${VAD_TEST} PROPERTIES LABELS "unit")
@@ -106,7 +106,7 @@ target_include_directories(${VAD_TEST} PRIVATE ../include ../ggml/include ../exa
 target_link_libraries(${VAD_TEST} PRIVATE common)
 target_compile_definitions(${VAD_TEST} PRIVATE
     WHISPER_MODEL_PATH="${PROJECT_SOURCE_DIR}/models/ggml-base.en.bin"
-    VAD_MODEL_PATH="${PROJECT_SOURCE_DIR}/models/for-tests-silero-v5.1.2-ggml.bin"
+    VAD_MODEL_PATH="${PROJECT_SOURCE_DIR}/models/for-tests-silero-v6.2.0-ggml.bin"
     SAMPLE_PATH="${PROJECT_SOURCE_DIR}/samples/jfk.wav")
 add_test(NAME ${VAD_TEST} COMMAND ${VAD_TEST})
 set_tests_properties(${VAD_TEST} PROPERTIES LABELS "base;en")

tests/test-vad-full.cpp

@@ -42,11 +42,13 @@ int main() {
     const int n_segments = whisper_full_n_segments(wctx);
     assert(n_segments == 1);
 
+
+    printf("Segment text:\n%s", whisper_full_get_segment_text(wctx, 0));
     assert(strcmp(" And so my fellow Americans, ask not what your country can do for you,"
                   " ask what you can do for your country.",
            whisper_full_get_segment_text(wctx, 0)) == 0);
-    assert(whisper_full_get_segment_t0(wctx, 0) == 29);
+    assert(whisper_full_get_segment_t0(wctx, 0) == 32);
-    assert(whisper_full_get_segment_t1(wctx, 0) == 1049);
+    assert(whisper_full_get_segment_t1(wctx, 0) == 1051);
 
     whisper_free(wctx);
 

tests/test-vad.cpp

@@ -36,7 +36,7 @@ struct whisper_vad_segments * test_detect_timestamps(
         struct whisper_vad_context * vctx,
         struct whisper_vad_params params) {
     struct whisper_vad_segments * timestamps = whisper_vad_segments_from_probs(vctx, params);
-    assert(whisper_vad_segments_n_segments(timestamps) == 5);
+    assert(whisper_vad_segments_n_segments(timestamps) == 4);
 
     for (int i = 0; i < whisper_vad_segments_n_segments(timestamps); ++i) {
         printf("VAD segment %d: start = %.2f, end = %.2f\n", i,