vad-speech-segments crash with default use-gpu set to true #3508
Description
I'm getting a crash when running just the vad-speech-segments.
I would like to integrate the vad-speech-segments in the audio.whisper R package. I noticed it does not crash when running without use-gpu = FALSE.
What is different between vad-speech-segments and when it's called through whisper-cli ?
whisper.cpp (master) $ cmake -B build -DGGML_CUDA=1 -DCMAKE_CUDA_ARCHITECTURES="75" -DWHISPER_BUILD_EXAMPLES=1
CMake Deprecation Warning at CMakeLists.txt:1 (cmake_minimum_required):
Compatibility with CMake < 3.10 will be removed from a future version of
CMake.
Update the VERSION argument <min> value. Or, use the <min>...<max> syntax
to tell CMake that the project requires at least <min> but has been updated
to work with policies introduced by <max> or earlier.
-- Warning: ccache not found - consider installing it for faster compilation or disable this warning with GGML_CCACHE=OFF
-- CMAKE_SYSTEM_PROCESSOR: x86_64
-- GGML_SYSTEM_ARCH: x86
-- Including CPU backend
-- x86 detected
-- Adding CPU backend variant ggml-cpu: -march=native
-- CUDA Toolkit found
-- Using CUDA architectures: 75
-- CUDA host compiler is GNU 11.4.0
-- Including CUDA backend
-- ggml version: 0.9.4
-- ggml commit: 999a7e0c
-- Configuring done (0.3s)
-- Generating done (0.1s)
-- Build files have been written to: /workspace/speech_to_text/whisper.cpp/build
whisper.cpp (master) $ cmake --build build -j 6 --config Release
[ 0%] Building C object ggml/src/CMakeFiles/ggml-base.dir/ggml.c.o
[ 1%] Building CXX object examples/deprecation-warning/CMakeFiles/main.dir/deprecation-warning.cpp.o
[ 2%] Building CXX object ggml/src/CMakeFiles/ggml-base.dir/ggml.cpp.o
[ 2%] Building CXX object examples/deprecation-warning/CMakeFiles/bench.dir/deprecation-warning.cpp.o
[ 3%] Building C object ggml/src/CMakeFiles/ggml-base.dir/ggml-alloc.c.o
[ 3%] Building CXX object ggml/src/CMakeFiles/ggml-base.dir/ggml-backend.cpp.o
[ 4%] Building CXX object ggml/src/CMakeFiles/ggml-base.dir/ggml-opt.cpp.o
[ 5%] Linking CXX executable ../../bin/bench
[ 5%] Linking CXX executable ../../bin/main
[ 5%] Built target bench
[ 5%] Built target main
[ 5%] Building CXX object ggml/src/CMakeFiles/ggml-base.dir/ggml-threading.cpp.o
[ 6%] Building C object ggml/src/CMakeFiles/ggml-base.dir/ggml-quants.c.o
[ 7%] Building CXX object ggml/src/CMakeFiles/ggml-base.dir/gguf.cpp.o
[ 7%] Linking CXX shared library libggml-base.so
[ 7%] Built target ggml-base
[ 8%] Building C object ggml/src/CMakeFiles/ggml-cpu.dir/ggml-cpu/ggml-cpu.c.o
[ 8%] Building CXX object ggml/src/CMakeFiles/ggml-cpu.dir/ggml-cpu/ggml-cpu.cpp.o
[ 9%] Building CXX object ggml/src/CMakeFiles/ggml-cpu.dir/ggml-cpu/repack.cpp.o
[ 9%] Building CXX object ggml/src/CMakeFiles/ggml-cpu.dir/ggml-cpu/hbm.cpp.o
[ 10%] Building C object ggml/src/CMakeFiles/ggml-cpu.dir/ggml-cpu/quants.c.o
[ 11%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/acc.cu.o
[ 12%] Building CXX object ggml/src/CMakeFiles/ggml-cpu.dir/ggml-cpu/traits.cpp.o
[ 12%] Building CXX object ggml/src/CMakeFiles/ggml-cpu.dir/ggml-cpu/amx/amx.cpp.o
[ 13%] Building CXX object ggml/src/CMakeFiles/ggml-cpu.dir/ggml-cpu/amx/mmq.cpp.o
[ 13%] Building CXX object ggml/src/CMakeFiles/ggml-cpu.dir/ggml-cpu/binary-ops.cpp.o
[ 14%] Building CXX object ggml/src/CMakeFiles/ggml-cpu.dir/ggml-cpu/unary-ops.cpp.o
[ 15%] Building CXX object ggml/src/CMakeFiles/ggml-cpu.dir/ggml-cpu/vec.cpp.o
[ 15%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/add-id.cu.o
[ 15%] Building CXX object ggml/src/CMakeFiles/ggml-cpu.dir/ggml-cpu/ops.cpp.o
[ 16%] Building C object ggml/src/CMakeFiles/ggml-cpu.dir/ggml-cpu/arch/x86/quants.c.o
[ 17%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/arange.cu.o
[ 17%] Building CXX object ggml/src/CMakeFiles/ggml-cpu.dir/ggml-cpu/arch/x86/repack.cpp.o
[ 17%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/argmax.cu.o
[ 18%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/argsort.cu.o
[ 18%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/binbcast.cu.o
[ 19%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/clamp.cu.o
[ 20%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/concat.cu.o
[ 20%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/conv-transpose-1d.cu.o
[ 21%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/conv2d-dw.cu.o
[ 21%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/conv2d-transpose.cu.o
[ 22%] Linking CXX shared library libggml-cpu.so
[ 22%] Built target ggml-cpu
[ 23%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/conv2d.cu.o
[ 24%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/convert.cu.o
[ 24%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/count-equal.cu.o
[ 25%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/cpy.cu.o
[ 25%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/cross-entropy-loss.cu.o
[ 26%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/diagmask.cu.o
[ 27%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/fattn-tile.cu.o
[ 27%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/fattn-wmma-f16.cu.o
[ 28%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/fattn.cu.o
[ 28%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/getrows.cu.o
[ 29%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/ggml-cuda.cu.o
[ 29%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/gla.cu.o
[ 30%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/im2col.cu.o
[ 31%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/mean.cu.o
[ 31%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/mmf.cu.o
[ 32%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/mmid.cu.o
[ 32%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/mmq.cu.o
[ 33%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/mmvf.cu.o
[ 34%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/mmvq.cu.o
[ 34%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/norm.cu.o
[ 35%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/opt-step-adamw.cu.o
[ 35%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/opt-step-sgd.cu.o
[ 36%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/out-prod.cu.o
[ 37%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/pad.cu.o
[ 37%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/pad_reflect_1d.cu.o
[ 38%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/pool2d.cu.o
[ 38%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/quantize.cu.o
[ 39%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/roll.cu.o
[ 39%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/rope.cu.o
[ 40%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/scale.cu.o
[ 41%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/set-rows.cu.o
[ 41%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/softcap.cu.o
[ 42%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/softmax.cu.o
[ 42%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/ssm-conv.cu.o
[ 43%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/ssm-scan.cu.o
[ 44%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/sum.cu.o
[ 44%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/sumrows.cu.o
[ 45%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/topk-moe.cu.o
[ 45%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/tsembd.cu.o
[ 46%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/unary.cu.o
[ 47%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/upscale.cu.o
[ 47%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/wkv.cu.o
[ 48%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/template-instances/fattn-tile-instance-dkq112-dv112.cu.o
[ 48%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/template-instances/fattn-tile-instance-dkq128-dv128.cu.o
[ 49%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/template-instances/fattn-tile-instance-dkq256-dv256.cu.o
[ 49%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/template-instances/fattn-tile-instance-dkq40-dv40.cu.o
[ 50%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/template-instances/fattn-tile-instance-dkq576-dv512.cu.o
[ 51%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/template-instances/fattn-tile-instance-dkq64-dv64.cu.o
[ 51%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/template-instances/fattn-tile-instance-dkq80-dv80.cu.o
[ 52%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/template-instances/fattn-tile-instance-dkq96-dv96.cu.o
[ 52%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/template-instances/fattn-mma-f16-instance-ncols1_1-ncols2_16.cu.o
[ 53%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/template-instances/fattn-mma-f16-instance-ncols1_1-ncols2_8.cu.o
[ 54%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/template-instances/fattn-mma-f16-instance-ncols1_16-ncols2_1.cu.o
[ 54%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/template-instances/fattn-mma-f16-instance-ncols1_16-ncols2_2.cu.o
[ 55%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/template-instances/fattn-mma-f16-instance-ncols1_16-ncols2_4.cu.o
[ 55%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/template-instances/fattn-mma-f16-instance-ncols1_2-ncols2_16.cu.o
[ 56%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/template-instances/fattn-mma-f16-instance-ncols1_2-ncols2_4.cu.o
[ 57%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/template-instances/fattn-mma-f16-instance-ncols1_2-ncols2_8.cu.o
[ 57%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/template-instances/fattn-mma-f16-instance-ncols1_32-ncols2_1.cu.o
[ 58%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/template-instances/fattn-mma-f16-instance-ncols1_32-ncols2_2.cu.o
[ 58%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/template-instances/fattn-mma-f16-instance-ncols1_4-ncols2_16.cu.o
[ 59%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/template-instances/fattn-mma-f16-instance-ncols1_4-ncols2_2.cu.o
[ 59%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/template-instances/fattn-mma-f16-instance-ncols1_4-ncols2_4.cu.o
[ 60%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/template-instances/fattn-mma-f16-instance-ncols1_4-ncols2_8.cu.o
[ 61%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/template-instances/fattn-mma-f16-instance-ncols1_64-ncols2_1.cu.o
[ 61%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/template-instances/fattn-mma-f16-instance-ncols1_8-ncols2_1.cu.o
[ 62%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/template-instances/fattn-mma-f16-instance-ncols1_8-ncols2_2.cu.o
[ 62%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/template-instances/fattn-mma-f16-instance-ncols1_8-ncols2_4.cu.o
[ 63%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/template-instances/fattn-mma-f16-instance-ncols1_8-ncols2_8.cu.o
[ 64%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/template-instances/mmq-instance-iq1_s.cu.o
[ 64%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/template-instances/mmq-instance-iq2_s.cu.o
[ 65%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/template-instances/mmq-instance-iq2_xs.cu.o
[ 65%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/template-instances/mmq-instance-iq2_xxs.cu.o
[ 66%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/template-instances/mmq-instance-iq3_s.cu.o
[ 67%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/template-instances/mmq-instance-iq3_xxs.cu.o
[ 67%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/template-instances/mmq-instance-iq4_nl.cu.o
[ 68%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/template-instances/mmq-instance-iq4_xs.cu.o
[ 68%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/template-instances/mmq-instance-mxfp4.cu.o
[ 69%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/template-instances/mmq-instance-q2_k.cu.o
[ 69%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/template-instances/mmq-instance-q3_k.cu.o
[ 70%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/template-instances/mmq-instance-q4_0.cu.o
[ 71%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/template-instances/mmq-instance-q4_1.cu.o
[ 71%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/template-instances/mmq-instance-q4_k.cu.o
[ 72%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/template-instances/mmq-instance-q5_0.cu.o
[ 72%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/template-instances/mmq-instance-q5_1.cu.o
[ 73%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/template-instances/mmq-instance-q5_k.cu.o
[ 74%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/template-instances/mmq-instance-q6_k.cu.o
[ 74%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/template-instances/mmq-instance-q8_0.cu.o
[ 75%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/template-instances/mmf-instance-ncols_1.cu.o
[ 75%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/template-instances/mmf-instance-ncols_10.cu.o
[ 76%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/template-instances/mmf-instance-ncols_11.cu.o
[ 77%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/template-instances/mmf-instance-ncols_12.cu.o
[ 77%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/template-instances/mmf-instance-ncols_13.cu.o
[ 78%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/template-instances/mmf-instance-ncols_14.cu.o
[ 78%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/template-instances/mmf-instance-ncols_15.cu.o
[ 79%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/template-instances/mmf-instance-ncols_16.cu.o
[ 79%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/template-instances/mmf-instance-ncols_2.cu.o
[ 80%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/template-instances/mmf-instance-ncols_3.cu.o
[ 81%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/template-instances/mmf-instance-ncols_4.cu.o
[ 81%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/template-instances/mmf-instance-ncols_5.cu.o
[ 82%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/template-instances/mmf-instance-ncols_6.cu.o
[ 82%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/template-instances/mmf-instance-ncols_7.cu.o
[ 83%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/template-instances/mmf-instance-ncols_8.cu.o
[ 84%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/template-instances/mmf-instance-ncols_9.cu.o
[ 84%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/template-instances/fattn-vec-instance-q4_0-q4_0.cu.o
[ 85%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/template-instances/fattn-vec-instance-q8_0-q8_0.cu.o
[ 85%] Building CUDA object ggml/src/ggml-cuda/CMakeFiles/ggml-cuda.dir/template-instances/fattn-vec-instance-f16-f16.cu.o
[ 86%] Linking CUDA shared library libggml-cuda.so
[ 86%] Built target ggml-cuda
[ 86%] Building CXX object ggml/src/CMakeFiles/ggml.dir/ggml-backend-reg.cpp.o
[ 87%] Linking CXX shared library libggml.so
[ 87%] Built target ggml
[ 87%] Building CXX object src/CMakeFiles/whisper.dir/whisper.cpp.o
[ 88%] Linking CXX shared library libwhisper.so
[ 88%] Built target whisper
[ 88%] Building CXX object examples/CMakeFiles/common.dir/common.cpp.o
[ 90%] Building CXX object examples/CMakeFiles/common.dir/common-ggml.cpp.o
[ 90%] Building CXX object examples/bench/CMakeFiles/whisper-bench.dir/bench.cpp.o
[ 90%] Building CXX object examples/CMakeFiles/common.dir/common-whisper.cpp.o
[ 91%] Building CXX object examples/CMakeFiles/common.dir/grammar-parser.cpp.o
[ 91%] Linking CXX executable ../../bin/whisper-bench
[ 91%] Built target whisper-bench
[ 92%] Linking CXX static library libcommon.a
[ 92%] Built target common
[ 93%] Building CXX object tests/CMakeFiles/test-vad.dir/test-vad.cpp.o
[ 95%] Building CXX object tests/CMakeFiles/test-vad-full.dir/test-vad-full.cpp.o
[ 95%] Building CXX object examples/cli/CMakeFiles/whisper-cli.dir/cli.cpp.o
[ 96%] Building CXX object examples/server/CMakeFiles/whisper-server.dir/server.cpp.o
[ 97%] Building CXX object examples/quantize/CMakeFiles/quantize.dir/quantize.cpp.o
[ 97%] Building CXX object examples/vad-speech-segments/CMakeFiles/vad-speech-segments.dir/speech.cpp.o
[ 98%] Linking CXX executable ../bin/test-vad-full
[ 98%] Linking CXX executable ../bin/test-vad
[ 98%] Built target test-vad-full
[ 98%] Built target test-vad
[ 99%] Linking CXX executable ../../bin/vad-speech-segments
[ 99%] Built target vad-speech-segments
[ 99%] Linking CXX executable ../../bin/quantize
[ 99%] Built target quantize
[ 99%] Linking CXX executable ../../bin/whisper-cli
[ 99%] Built target whisper-cli
[100%] Linking CXX executable ../../bin/whisper-server
[100%] Built target whisper-server
whisper.cpp (master) $ ./build/bin/vad-speech-segments -vm /workspace/speech_to_text/whisper.cpp/models/ggml-silero-v5.1.2.bin --threads 4 --use-gpu -f example.wav
whisper_vad_init_from_file_with_params: loading VAD model from '/workspace/speech_to_text/whisper.cpp/models/ggml-silero-v5.1.2.bin'
whisper_vad_init_with_params: model type: silero-16k
whisper_vad_init_with_params: model version: 5.1.2
whisper_vad_init_with_params: n_encoder_layers = 4
whisper_vad_init_with_params: encoder_in_channels[0] = 129
whisper_vad_init_with_params: encoder_in_channels[1] = 128
whisper_vad_init_with_params: encoder_in_channels[2] = 64
whisper_vad_init_with_params: encoder_in_channels[3] = 64
whisper_vad_init_with_params: encoder_out_channels[0] = 128
whisper_vad_init_with_params: encoder_out_channels[1] = 64
whisper_vad_init_with_params: encoder_out_channels[2] = 64
whisper_vad_init_with_params: encoder_out_channels[3] = 128
whisper_vad_init_with_params: lstm_input_size = 128
whisper_vad_init_with_params: lstm_hidden_size = 128
whisper_vad_init_with_params: final_conv_in = 128
whisper_vad_init_with_params: final_conv_out = 1
ggml_cuda_init: GGML_CUDA_FORCE_MMQ: no
ggml_cuda_init: GGML_CUDA_FORCE_CUBLAS: no
ggml_cuda_init: found 1 CUDA devices:
Device 0: Tesla T4, compute capability 7.5, VMM: yes
whisper_vad_init_with_params: CUDA0 total size = 0.88 MB
whisper_vad_init_with_params: model size = 0.88 MB
whisper_backend_init_gpu: no GPU found
/workspace/speech_to_text/whisper.cpp/ggml/src/ggml-backend.cpp:796: pre-allocated tensor (leaf_0) in a buffer (CUDA0) that cannot run the operation (NONE)
[New LWP 45516]
[New LWP 45523]
[Thread debugging using libthread_db enabled]
Using host libthread_db library "/lib/x86_64-linux-gnu/libthread_db.so.1".
0x00007f7604b9642f in wait4 () from /lib/x86_64-linux-gnu/libc.so.6
#0 0x00007f7604b9642f in wait4 () from /lib/x86_64-linux-gnu/libc.so.6
#1 0x00007f7604a212eb in ggml_print_backtrace () from /workspace/speech_to_text/whisper.cpp/build/ggml/src/libggml-base.so
#2 0x00007f7604a21482 in ggml_abort () from /workspace/speech_to_text/whisper.cpp/build/ggml/src/libggml-base.so
#3 0x00007f7604a38974 in ggml_backend_sched_backend_id_from_cur(ggml_backend_sched*, ggml_tensor*) () from /workspace/speech_to_text/whisper.cpp/build/ggml/src/libggml-base.so
#4 0x00007f7604a3b02f in ggml_backend_sched_split_graph () from /workspace/speech_to_text/whisper.cpp/build/ggml/src/libggml-base.so
#5 0x00007f7604a3bf7d in ggml_backend_sched_alloc_graph () from /workspace/speech_to_text/whisper.cpp/build/ggml/src/libggml-base.so
#6 0x00007f76050561a3 in whisper_sched_graph_init(whisper_sched&, std::vector<ggml_backend*, std::allocator<ggml_backend*> >, std::function<ggml_cgraph* ()>&&) [clone .isra.0] () from /workspace/speech_to_text/whisper.cpp/build/src/libwhisper.so.1
#7 0x00007f760505fff6 in whisper_vad_init_with_params () from /workspace/speech_to_text/whisper.cpp/build/src/libwhisper.so.1
#8 0x00007f7605060637 in whisper_vad_init_from_file_with_params () from /workspace/speech_to_text/whisper.cpp/build/src/libwhisper.so.1
#9 0x0000556334650758 in main ()
[Inferior 1 (process 45515) detached]
Aborted
While integrated with whisper all runs fine:
whisper.cpp (master) $ ./build/bin/whisper-cli -vm /workspace/speech_to_text/whisper.cpp/models/ggml-silero-v5.1.2.bin --vad -f example.wav -m models/ggml-large-v3-turbo-q5_0.bin
whisper_init_from_file_with_params_no_state: loading model from 'models/ggml-large-v3-turbo-q5_0.bin'
whisper_init_with_params_no_state: use gpu = 1
whisper_init_with_params_no_state: flash attn = 1
whisper_init_with_params_no_state: gpu_device = 0
whisper_init_with_params_no_state: dtw = 0
ggml_cuda_init: GGML_CUDA_FORCE_MMQ: no
ggml_cuda_init: GGML_CUDA_FORCE_CUBLAS: no
ggml_cuda_init: found 1 CUDA devices:
Device 0: Tesla T4, compute capability 7.5, VMM: yes
whisper_init_with_params_no_state: devices = 2
whisper_init_with_params_no_state: backends = 2
whisper_model_load: loading model
whisper_model_load: n_vocab = 51866
whisper_model_load: n_audio_ctx = 1500
whisper_model_load: n_audio_state = 1280
whisper_model_load: n_audio_head = 20
whisper_model_load: n_audio_layer = 32
whisper_model_load: n_text_ctx = 448
whisper_model_load: n_text_state = 1280
whisper_model_load: n_text_head = 20
whisper_model_load: n_text_layer = 4
whisper_model_load: n_mels = 128
whisper_model_load: ftype = 8
whisper_model_load: qntvr = 2
whisper_model_load: type = 5 (large v3)
whisper_model_load: adding 1609 extra tokens
whisper_model_load: n_langs = 100
whisper_model_load: CUDA0 total size = 573.45 MB
whisper_model_load: model size = 573.40 MB
whisper_backend_init_gpu: device 0: CUDA0 (type: 1)
whisper_backend_init_gpu: found GPU device 0: CUDA0 (type: 1, cnt: 0)
whisper_backend_init_gpu: using CUDA0 backend
whisper_init_state: kv self size = 10.49 MB
whisper_init_state: kv cross size = 31.46 MB
whisper_init_state: kv pad size = 7.86 MB
whisper_init_state: compute buffer (conv) = 37.69 MB
whisper_init_state: compute buffer (encode) = 55.35 MB
whisper_init_state: compute buffer (cross) = 9.27 MB
whisper_init_state: compute buffer (decode) = 100.04 MB
system_info: n_threads = 4 / 32 | WHISPER : COREML = 0 | OPENVINO = 0 | CUDA : ARCHS = 750 | PEER_MAX_BATCH_SIZE = 128 | CPU : SSE3 = 1 | SSSE3 = 1 | AVX = 1 | AVX2 = 1 | F16C = 1 | FMA = 1 | BMI2 = 1 | AVX512 = 1 | AVX512_VNNI = 1 | OPENMP = 1 | REPACK = 1 |
main: processing 'example.wav' (4802560 samples, 300.2 sec), 4 threads, 1 processors, 5 beams + best of 5, lang = en, task = transcribe, timestamps = 1 ...
whisper_full: VAD is enabled, processing speech segments only
whisper_vad: VAD is enabled, processing speech segments only
whisper_vad_init_from_file_with_params: loading VAD model from '/workspace/speech_to_text/whisper.cpp/models/ggml-silero-v5.1.2.bin'
whisper_vad_init_with_params: model type: silero-16k
whisper_vad_init_with_params: model version: 5.1.2
whisper_vad_init_with_params: n_encoder_layers = 4
whisper_vad_init_with_params: encoder_in_channels[0] = 129
whisper_vad_init_with_params: encoder_in_channels[1] = 128
whisper_vad_init_with_params: encoder_in_channels[2] = 64
whisper_vad_init_with_params: encoder_in_channels[3] = 64
whisper_vad_init_with_params: encoder_out_channels[0] = 128
whisper_vad_init_with_params: encoder_out_channels[1] = 64
whisper_vad_init_with_params: encoder_out_channels[2] = 64
whisper_vad_init_with_params: encoder_out_channels[3] = 128
whisper_vad_init_with_params: lstm_input_size = 128
whisper_vad_init_with_params: lstm_hidden_size = 128
whisper_vad_init_with_params: final_conv_in = 128
whisper_vad_init_with_params: final_conv_out = 1
whisper_vad_init_with_params: CPU total size = 0.88 MB
whisper_vad_init_with_params: model size = 0.88 MB
whisper_backend_init_gpu: no GPU found
whisper_vad_init_context: compute buffer (VAD) = 1.60 MB
whisper_vad_segments_from_samples: detecting speech timestamps in 4802560 samples
whisper_vad_detect_speech: detecting speech in 4802560 samples
whisper_vad_detect_speech: n_chunks: 9380
whisper_vad_detect_speech: props size: 9380
whisper_vad_detect_speech: vad time = 1668.74 ms processing 4802560 samples
whisper_vad_segments_from_probs: detecting speech timestamps using 9380 probabilities
whisper_vad_segments_from_probs: Merged 5 adjacent segments, now have 7 segments
whisper_vad_segments_from_probs: Final speech segments after filtering: 7
whisper_vad_segments_from_probs: VAD segment 0: start = 0.10, end = 0.70 (duration: 0.60)
whisper_vad_segments_from_probs: VAD segment 1: start = 1.15, end = 1.92 (duration: 0.77)
whisper_vad_segments_from_probs: VAD segment 2: start = 2.75, end = 4.13 (duration: 1.38)
whisper_vad_segments_from_probs: VAD segment 3: start = 4.39, end = 5.98 (duration: 1.59)
whisper_vad_segments_from_probs: VAD segment 4: start = 6.18, end = 89.44 (duration: 83.26)
whisper_vad_segments_from_probs: VAD segment 5: start = 89.60, end = 191.68 (duration: 102.08)
whisper_vad_segments_from_probs: VAD segment 6: start = 191.84, end = 300.16 (duration: 108.32)
whisper_vad: detected 7 speech segments
whisper_vad: Including segment 0: 0.10 - 0.80 (duration: 0.70)
whisper_vad: Including segment 1: 1.15 - 2.02 (duration: 0.87)
whisper_vad: Including segment 2: 2.75 - 4.23 (duration: 1.48)
whisper_vad: Including segment 3: 4.39 - 6.08 (duration: 1.69)
whisper_vad: Including segment 4: 6.18 - 89.54 (duration: 83.36)
whisper_vad: Including segment 5: 89.60 - 191.78 (duration: 102.18)
whisper_vad: Including segment 6: 191.84 - 300.16 (duration: 108.32)
whisper_vad: total duration of speech segments: 298.60 seconds
whisper_vad: vad_segment_info: orig_start: 0.10, orig_end: 0.70, vad_start: 0.00, vad_end: 0.70
whisper_vad: vad_segment_info: orig_start: 1.15, orig_end: 1.92, vad_start: 0.80, vad_end: 1.67
whisper_vad: vad_segment_info: orig_start: 2.75, orig_end: 4.13, vad_start: 1.77, vad_end: 3.25
whisper_vad: vad_segment_info: orig_start: 4.39, orig_end: 5.98, vad_start: 3.35, vad_end: 5.04
whisper_vad: vad_segment_info: orig_start: 6.18, orig_end: 89.44, vad_start: 5.14, vad_end: 88.50
whisper_vad: vad_segment_info: orig_start: 89.60, orig_end: 191.68, vad_start: 88.60, vad_end: 190.78
whisper_vad: vad_segment_info: orig_start: 191.84, orig_end: 300.16, vad_start: 190.88, vad_end: 299.20
whisper_vad: Created time mapping table with 1491 points
whisper_vad: Reduced audio from 4802560 to 4787200 samples (0.3% reduction)
[00:00:00.100 --> 00:00:01.620] Look at the picture.
[00:00:01.620 --> 00:00:03.860] See the skull.
[00:00:03.860 --> 00:00:05.800] The part of bone removed.
[00:00:05.800 --> 00:00:08.810] The master race Frankenstein radio controls.
[00:00:08.810 --> 00:00:10.990] The brain thoughts broadcasting radio.
[00:00:10.990 --> 00:00:12.350] The eyesight television.
[00:00:12.350 --> 00:00:14.270] The Frankenstein earphone radio.
[00:00:14.270 --> 00:00:16.200] The threshold brainwash radio.
[00:00:16.200 --> 00:00:21.280] The latest new skull reforming to contain all Frankenstein controls.
[00:00:21.280 --> 00:00:24.610] Even in thin skulls of white pedigree males.
[00:00:24.610 --> 00:00:26.990] Visible Frankenstein controls.
[00:00:26.990 --> 00:00:30.230] The synthetic nerve radio directional antenna loop.
[00:00:30.630 --> 00:00:31.880] Make copies for yourself.
[00:00:31.880 --> 00:00:36.740] There is no escape from this worse gangster police state.
[00:00:36.740 --> 00:00:39.950] Using all of the deadly gangster Frankenstein controls.
[00:00:39.950 --> 00:00:44.290] In 1965 CIA gangster police beat me bloody.
[00:00:44.290 --> 00:00:47.110] Dragged me in chains from Kennedy New York airport.
[00:00:47.110 --> 00:00:50.200] Since then I hide in forced jobless poverty.
[00:00:50.200 --> 00:00:53.960] Isolated alone in this low deadly nigger town old house.
[00:00:53.960 --> 00:00:57.890] The brazen deadly gangster police and nigger puppet underlings.
[00:00:57.890 --> 00:00:59.970] Spray me with poison nerve gas.
[00:01:00.210 --> 00:01:02.370] From automobile exhausts and even lawn mowers.
[00:01:02.370 --> 00:01:04.440] Deadly assaults even in my yard.
[00:01:04.440 --> 00:01:05.380] With knives.
[00:01:05.380 --> 00:01:06.960] Even bricks and stones.
[00:01:06.960 --> 00:01:10.120] Even deadly touched tabin or electric shock flashlights.
[00:01:10.120 --> 00:01:15.650] Even remote electronically controlled around corners projection of deadly touched tarantula spiders.
[00:01:16.030 --> 00:01:18.410] Or even bloody murder accidents.
[00:01:18.410 --> 00:01:21.700] To shut me up forever with a sneak undetectable extermination.
[00:01:21.700 --> 00:01:26.720] Even with trained parroting puppet assassins in maximum security insanity prison.
[00:01:26.720 --> 00:01:29.120] For writing these unforgivable truths.
[00:01:29.120 --> 00:01:32.130] Until my undetectable extermination eye.
[00:01:32.130 --> 00:01:34.610] I, Francis E. Deck Esquire.
[00:01:34.610 --> 00:01:35.610] 29 Maple Avenue.
[00:01:35.610 --> 00:01:45.120] I stand alone against your mad, deadly, worldwide, conspiratorial gangster computer god communism.
[00:01:45.120 --> 00:01:48.280] With wall-to-wall deadly gangster protection.
[00:01:48.280 --> 00:01:53.190] Life-long sworn conspirators murder incorporated organized crime.
[00:01:53.190 --> 00:01:54.350] The police and judges.
[00:01:54.350 --> 00:02:00.380] The deadly sneak parroting puppet gangsters using all the gangster deadly Frankenstein controls.
[00:02:00.380 --> 00:02:03.160] These hangman rope sneak deadly gangsters.
[00:02:03.160 --> 00:02:15.650] The judges and the police trick, trap, rob, wreck, butcher, and murder the people to keep them terrorized in gangster Frankenstein earphone radio slavery for the communist gangster government.
[00:02:15.650 --> 00:02:19.990] And con artist parroting puppet gangster playboy scum on top.
[00:02:19.990 --> 00:02:25.500] The secret work of all police in order to maintain a communist closed society.
[00:02:25.500 --> 00:02:36.390] The same worldwide mad, deadly, communist, gangster, computer god that controls you as a terrorized gangster Frankenstein earphone radio slave parroting puppet.
[00:02:36.390 --> 00:02:41.700] You are a terrorized member of the master race worldwide.
[00:02:41.700 --> 00:02:48.600] Four billion eyesight, television camera, guinea pig, communist gangster, computer god, master race.
[00:02:48.600 --> 00:02:59.490] You are living, thinking, mad, deadly, world-wide, communist, gangster, computer god, secret, overall plan, world-wide, living, death, Frankenstein slavery.
[00:02:59.490 --> 00:03:04.880] To explore and control the entire universe with the endless stairway to the stars.
[00:03:04.880 --> 00:03:11.480] Namely, the man-made, inside-out planets with nucleonic powered speeds much faster than the speed of light.
[00:03:11.480 --> 00:03:16.460] Look up and see the gangster computer god concocted new fake starry sky.
[00:03:16.460 --> 00:03:24.440] The worldwide completely controlled deadly degenerative climate and atmosphere through the new world-round translucent exotic gaseous envelope.
[00:03:24.440 --> 00:03:31.440] Which the worldwide communist gangster computer god manipulates through countless exactly positioned satellites.
[00:03:32.300 --> 00:03:44.200] The new fake, phony stars in the synthetic sky for ages before Frankenstein controls, apointic niggers, interpretable with apes, had no alphabet, not even numerals.
[00:03:44.200 --> 00:03:55.200] Slavery conspiracy over 300 years ago, ideally tiny-brained, apointic nigger, gangster government eyesight, TV gangster, spy cameras, computer god, new world order degeneration.
[00:03:55.200 --> 00:04:09.400] With gifted, with all gangster Frankenstein controls, nigger deadly gangster parroting puppets or nigger brain programmed robots, deadly ape Frankenstein machines, degenerative disease to eternal Frankenstein slavery.
[00:04:09.400 --> 00:04:20.600] Overall plan through one world communism, top secret code word, meaning worldwide, absolutely helpless and hopeless, simple language, mongrel, mulatto, apointic niggers.
[00:04:20.600 --> 00:04:34.000] Worldwide systematic instant plastic surgery, butchery murder, fake aging so all people are dead or useless by age 70 done at night to you as a Frankenstein slave parroting puppet gangster slave.
[00:04:34.000 --> 00:04:41.300] Now even you know I am a menace to your worldwide mad, deadly communist gangster computer god.
[00:04:41.300 --> 00:04:53.700] Therefore, I must go to extermination before I am exterminated by this gangster computer god concocted and controlled, worst mongrel, organized crime, murder incorporated gangster communist government.
[00:04:53.700 --> 00:05:00.160] I hand you the secrets to save the entire human race and the entire human race.
whisper_print_timings: load time = 538.78 ms
whisper_print_timings: fallbacks = 0 p / 0 h
whisper_print_timings: mel time = 322.52 ms
whisper_print_timings: sample time = 2553.22 ms / 5036 runs ( 0.51 ms per run)
whisper_print_timings: encode time = 3217.73 ms / 16 runs ( 201.11 ms per run)
whisper_print_timings: decode time = 101.53 ms / 74 runs ( 1.37 ms per run)
whisper_print_timings: batchd time = 1944.80 ms / 4885 runs ( 0.40 ms per run)
whisper_print_timings: prompt time = 235.48 ms / 3289 runs ( 0.07 ms per run)
whisper_print_timings: total time = 10887.71 ms