Remove single file deployment for streaming instantiation and smaller size.

This commit is contained in:
msqr1
2024-08-28 19:15:56 -07:00
parent 01068a2c6c
commit 0166c153bd
20 changed files with 87 additions and 86 deletions

View File

@@ -4,15 +4,15 @@ MAX_THREADS=${MAX_THREADS:-1}
EMSDK=${EMSDK:-../emsdk}
JOBS=${JOBS:-$(nproc)}
if [ $EMSDK != ../emsdk ] && [ ! -d $EMSDK ]; then
if [ "$EMSDK" != ../emsdk ] && [ ! -d "$EMSDK" ]; then
echo "Invalid emsdk path"
exit 1
fi
if [ $MAX_THREADS -lt 1 ]; then
fi
if [ "$MAX_THREADS" -lt 1 ]; then
echo "MAX_THREADS must be greater than or equal to 1"
exit 1
fi
if [ $JOBS -lt 1 ]; then
if [ "$JOBS" -lt 1 ]; then
echo "JOBS must be greater than or equal to 1"
exit 1
fi
@@ -20,15 +20,15 @@ if ! [[ $INITIAL_MEMORY =~ ^[0-9]+([kmgt]b)?$ ]]; then
echo "INITIAL_MEMORY valid suffixes are kb, mb, gb, tb, none (bytes)"
exit 1
fi
if [ $EMSDK = ../emsdk ] && [ ! -d $EMSDK ]; then
if [ "$EMSDK" = ../emsdk ] && [ ! -d "$EMSDK" ]; then
echo "Installing emsdk + Emscripten..."
git clone --depth=1 https://github.com/emscripten-core/emsdk.git ../emsdk &&
cd ../emsdk &&
./emsdk install 3.1.65 &&
./emsdk activate 3.1.65
fi
. $(realpath $EMSDK)/emsdk_env.sh &&
export PATH=:$PATH:$(realpath $EMSDK)/upstream/bin &&
. $(realpath "$EMSDK")/emsdk_env.sh &&
export PATH=:$PATH:$(realpath "$EMSDK")/upstream/bin &&
cd .. &&
SRC=$(realpath src)
@@ -37,47 +37,48 @@ VOSK=$(realpath vosk)
OPENFST=$(realpath openfst)
CLAPACK_WASM=$(realpath clapack-wasm)
if [ ! -d $OPENFST ]; then
if [ ! -d "$OPENFST" ]; then
rm -rf /tmp/openfst &&
git clone --depth=1 https://github.com/alphacep/openfst /tmp/openfst &&
cd /tmp/openfst &&
git apply $SRC/Openfst.patch
git apply "$SRC"/Openfst.patch
autoreconf -is &&
CXXFLAGS="-r -O3 -flto -msimd128 -mreference-types -mnontrapping-fptoint -mextended-const -msign-ext -mmutable-globals" LDFLAGS="-O3 -flto" emconfigure ./configure --prefix=$OPENFST --enable-static --disable-shared --enable-lookahead-fsts --enable-ngram-fsts --disable-bin &&
emmake make -j$JOBS install &&
echo "PACKAGE_VERSION = 1.8.0" >> $OPENFST/Makefile
CXXFLAGS="-r -O3 -flto -msimd128 -mreference-types -mnontrapping-fptoint -mextended-const -msign-ext -mmutable-globals" LDFLAGS="-O3 -flto" emconfigure ./configure --prefix="$OPENFST" --enable-static --disable-shared --enable-lookahead-fsts --enable-ngram-fsts --disable-bin &&
emmake make -j"$JOBS" install &&
echo "PACKAGE_VERSION = 1.8.0" >> "$OPENFST"/Makefile
fi
if [ ! -d $CLAPACK_WASM ]; then
git clone --depth=1 https://gitlab.inria.fr/multispeech/kaldi.web/clapack-wasm.git $CLAPACK_WASM &&
cd $CLAPACK_WASM &&
git apply $SRC/Clapack-wasm.patch &&
if [ ! -d "$CLAPACK_WASM" ]; then
git clone --depth=1 https://gitlab.inria.fr/multispeech/kaldi.web/clapack-wasm.git "$CLAPACK_WASM" &&
cd "$CLAPACK_WASM" &&
git apply "$SRC"/Clapack-wasm.patch &&
bash install_repo.sh emcc
fi
if [ ! -d $KALDI ]; then
git clone -b vosk --depth=1 https://github.com/alphacep/kaldi $KALDI &&
cd $KALDI/src &&
git apply $SRC/Kaldi.patch &&
CXXFLAGS="-O3 -UHAVE_EXECINFO_H -flto -msimd128 -mreference-types -mnontrapping-fptoint -mextended-const -msign-ext -mmutable-globals -Wno-unused-variable -Wno-unused-but-set-variable -g0" LDFLAGS="-O3 -lembind -flto -g0" emconfigure ./configure --use-cuda=no --with-cudadecoder=no --static --static-math=yes --static-fst=yes --debug-level=0 --fst-root=$OPENFST --clapack-root=$CLAPACK_WASM --host=WASM &&
emmake make -j$JOBS online2 rnnlm
if [ ! -d "$KALDI" ]; then
git clone -b vosk --depth=1 https://github.com/alphacep/kaldi "$KALDI" &&
cd "$KALDI"/src &&
git apply "$SRC"/Kaldi.patch &&
CXXFLAGS="-O3 -UHAVE_EXECINFO_H -flto -msimd128 -mreference-types -mnontrapping-fptoint -mextended-const -msign-ext -mmutable-globals -Wno-unused-variable -Wno-unused-but-set-variable -g0" LDFLAGS="-O3 -lembind -flto -g0" emconfigure ./configure --use-cuda=no --with-cudadecoder=no --static --static-math=yes --static-fst=yes --debug-level=0 --fst-root="$OPENFST" --clapack-root="$CLAPACK_WASM" --host=WASM &&
emmake make -j"$JOBS" online2 rnnlm
fi
if [ ! -d $VOSK ]; then
git clone -b v0.3.50 --depth=1 https://github.com/alphacep/vosk-api $VOSK &&
cd $VOSK/src &&
git apply $SRC/Vosk.patch &&
if [ ! -d "$VOSK" ]; then
git clone -b v0.3.50 --depth=1 https://github.com/alphacep/vosk-api "$VOSK" &&
cd "$VOSK"/src &&
git apply "$SRC"/Vosk.patch &&
VOSK_FILES="Recognizer.cc language_model.cc model.cc spk_model.cc vosk_api.cc" &&
em++ -O3 -flto -msimd128 -mreference-types -mnontrapping-fptoint -mextended-const -msign-ext -mmutable-globals -Wno-deprecated -I. -I$KALDI/src -I$OPENFST/include $VOSK_FILES -c &&
emar -rcs vosk.a ${VOSK_FILES//.cc/.o}
em++ -O3 -flto -msimd128 -mreference-types -mnontrapping-fptoint -mextended-const -msign-ext -mmutable-globals -Wno-deprecated -I. -I"$KALDI"/src -I"$OPENFST"/include "$VOSK_FILES" -c &&
emar -rcs vosk.a "${VOSK_FILES//.cc/.o}"
fi
cd $SRC &&
em++ Util.cc CommonModel.cc Recognizer.cc Bindings.cc -O3 -Wno-pthreads-mem-growth -DEMSCRIPTEN_HAS_UNBOUND_TYPE_NAMES=0 -fno-rtti -DMAX_THREADS=$MAX_THREADS -sWASMFS -sWASM_BIGINT -sSINGLE_FILE -sMODULARIZE -sEMBIND_STD_STRING_IS_UTF8 -sPTHREAD_POOL_DELAY_LOAD -sTEXTDECODER=2 -sPTHREAD_POOL_SIZE_STRICT=2 -sINITIAL_MEMORY=$INITIAL_MEMORY -sALLOW_MEMORY_GROWTH -sPTHREAD_POOL_SIZE=$MAX_THREADS -sPOLYFILL=0 -sEXIT_RUNTIME=0 -sINVOKE_RUN=0 -sSUPPORT_LONGJMP=0 -sALLOW_BLOCKING_ON_MAIN_THREAD=0 -sEXPORTED_FUNCTIONS=_malloc -sEXPORT_NAME=loadVosklet -sMALLOC=emmalloc -sEXPORTED_RUNTIME_METHODS=UTF8ToString,stringToUTF8OnStack -sENVIRONMENT=web,worker -I. -I$VOSK/src -L$KALDI/src -l:online2/kaldi-online2.a -l:decoder/kaldi-decoder.a -l:ivector/kaldi-ivector.a -l:gmm/kaldi-gmm.a -l:tree/kaldi-tree.a -l:feat/kaldi-feat.a -l:cudamatrix/kaldi-cudamatrix.a -l:lat/kaldi-lat.a -l:lm/kaldi-lm.a -l:rnnlm/kaldi-rnnlm.a -l:hmm/kaldi-hmm.a -l:nnet3/kaldi-nnet3.a -l:transform/kaldi-transform.a -l:matrix/kaldi-matrix.a -l:fstext/kaldi-fstext.a -l:util/kaldi-util.a -l:base/kaldi-base.a -L$OPENFST/lib -l:libfst.a -l:libfstngram.a -L$CLAPACK_WASM -l:CBLAS/lib/cblas.a -l:CLAPACK-3.2.1/lapack.a -l:CLAPACK-3.2.1/libcblaswr.a -l:f2c_BLAS-3.8.0/blas.a -l:libf2c/libf2c.a -L$VOSK/src -l:vosk.a -lembind -pthread -flto -msimd128 -mreference-types -mnontrapping-fptoint -mextended-const -msign-ext -mmutable-globals --pre-js Wrapper.js -o ../Vosklet.js &&
cd "$SRC" &&
em++ Util.cc CommonModel.cc Recognizer.cc Bindings.cc -O3 -Wno-pthreads-mem-growth -DEMSCRIPTEN_HAS_UNBOUND_TYPE_NAMES=0 -fno-rtti -DMAX_THREADS="$MAX_THREADS" -sWASMFS -sWASM_BIGINT -sMODULARIZE -sEMBIND_STD_STRING_IS_UTF8 -sPTHREAD_POOL_DELAY_LOAD -sTEXTDECODER=2 -sPTHREAD_POOL_SIZE_STRICT=2 -sINITIAL_MEMORY="$INITIAL_MEMORY" -sALLOW_MEMORY_GROWTH -sPTHREAD_POOL_SIZE="$MAX_THREADS" -sPOLYFILL=0 -sEXIT_RUNTIME=0 -sINVOKE_RUN=0 -sSUPPORT_LONGJMP=0 -sALLOW_BLOCKING_ON_MAIN_THREAD=0 -sEXPORTED_FUNCTIONS=_malloc -sEXPORT_NAME=loadVosklet -sMALLOC=emmalloc -sEXPORTED_RUNTIME_METHODS=UTF8ToString,stringToUTF8OnStack -sENVIRONMENT=web,worker -I. -I"$VOSK"/src -L"$KALDI"/src -l:online2/kaldi-online2.a -l:decoder/kaldi-decoder.a -l:ivector/kaldi-ivector.a -l:gmm/kaldi-gmm.a -l:tree/kaldi-tree.a -l:feat/kaldi-feat.a -l:cudamatrix/kaldi-cudamatrix.a -l:lat/kaldi-lat.a -l:lm/kaldi-lm.a -l:rnnlm/kaldi-rnnlm.a -l:hmm/kaldi-hmm.a -l:nnet3/kaldi-nnet3.a -l:transform/kaldi-transform.a -l:matrix/kaldi-matrix.a -l:fstext/kaldi-fstext.a -l:util/kaldi-util.a -l:base/kaldi-base.a -L"$OPENFST"/lib -l:libfst.a -l:libfstngram.a -L"$CLAPACK_WASM" -l:CBLAS/lib/cblas.a -l:CLAPACK-3.2.1/lapack.a -l:CLAPACK-3.2.1/libcblaswr.a -l:f2c_BLAS-3.8.0/blas.a -l:libf2c/libf2c.a -L"$VOSK"/src -l:vosk.a -lembind -pthread -flto -msimd128 -mreference-types -mnontrapping-fptoint -mextended-const -msign-ext -mmutable-globals --pre-js Wrapper.js -o ../Vosklet.js &&
cd .. &&
rm -f Vosklet.worker.js
cp Vosklet.js Examples/Vosklet.js &&
cp Vosklet.wasm Example/Vosklet.wasm &&
# Can't serve files from raw.githubusercontent with Content-Encoding: gzip header so the browser won't decompress automatically. Manually decompressing instead.
sed -i 's/res.body/new Response(res.body.pipeThrough(new DecompressionStream("gzip"))).body/' Examples/Vosklet.js &&