Support distil-small.en whisper (#472)
This commit is contained in:
Fangjun Kuang
2
.github/scripts/test-offline-whisper.sh
vendored
2
.github/scripts/test-offline-whisper.sh
vendored
@@ -22,6 +22,8 @@ tiny
|
||||
base
|
||||
small
|
||||
medium
|
||||
distil-medium.en
|
||||
distil-small.en
|
||||
)
|
||||
|
||||
for name in ${names[@]}; do
|
||||
|
||||
59
.github/workflows/export-whisper-to-onnx.yaml
vendored
59
.github/workflows/export-whisper-to-onnx.yaml
vendored
@@ -15,8 +15,9 @@ jobs:
|
||||
strategy:
|
||||
fail-fast: false
|
||||
matrix:
|
||||
os: [ubuntu-latest]
|
||||
model: ["distil-medium.en", "tiny.en", "base.en", "small.en", "medium.en", "tiny", "base", "small", "medium", "large", "large-v1", "large-v2"]
|
||||
os: [macos-latest]
|
||||
# model: ["distil-medium.en", "distil-small.en", "tiny.en", "base.en", "small.en", "medium.en", "tiny", "base", "small", "medium", "large", "large-v1", "large-v2", "distil-large-v2"]
|
||||
model: ["distil-medium.en", "distil-small.en", "tiny.en", "base.en", "small.en", "medium.en", "tiny", "base", "small", "medium"]
|
||||
python-version: ["3.8"]
|
||||
|
||||
steps:
|
||||
@@ -42,23 +43,33 @@ jobs:
|
||||
if [[ $model == distil-medium.en ]]; then
|
||||
wget -q -O distil-medium-en-original-model.bin https://huggingface.co/distil-whisper/distil-medium.en/resolve/main/original-model.bin
|
||||
ls -lh
|
||||
elif [[ $model == distil-large-v2 ]]; then
|
||||
wget -q -O distil-large-v2-original-model.bin https://huggingface.co/distil-whisper/distil-large-v2/resolve/main/original-model.bin
|
||||
ls -lh
|
||||
elif [[ $model == distil-small.en ]]; then
|
||||
wget -q -O distil-small-en-original-model.bin https://huggingface.co/distil-whisper/distil-small.en/resolve/main/original-model.bin
|
||||
ls -lh
|
||||
fi
|
||||
python3 ./export-onnx.py --model ${{ matrix.model }}
|
||||
# python3 -m onnxruntime.tools.convert_onnx_models_to_ort --optimization_style=Fixed ./
|
||||
|
||||
ls -lh
|
||||
|
||||
if [[ $model != distil-medium.en ]]; then
|
||||
ls -lh ~/.cache/whisper
|
||||
fi
|
||||
ls -lh ~/.cache/whisper || true
|
||||
ls -lh distil*original-model.bin || true
|
||||
rm -rf ~/.cache/whisper
|
||||
rm -f distil*original-model.bin
|
||||
|
||||
src=sherpa-onnx-whisper-${{ matrix.model }}
|
||||
|
||||
mkdir $src
|
||||
cp *.onnx $src/
|
||||
cp *tokens.txt $src
|
||||
cd ..
|
||||
mv whisper $src
|
||||
|
||||
echo "------------------------------"
|
||||
|
||||
cd $src
|
||||
du -h -d1 .
|
||||
ls -lh
|
||||
mkdir -p test_wavs
|
||||
cd test_wavs
|
||||
wget -q https://huggingface.co/csukuangfj/sherpa-onnx-whisper-medium.en/resolve/main/test_wavs/0.wav
|
||||
@@ -66,21 +77,32 @@ jobs:
|
||||
wget -q https://huggingface.co/csukuangfj/sherpa-onnx-whisper-medium.en/resolve/main/test_wavs/8k.wav
|
||||
wget -q https://huggingface.co/csukuangfj/sherpa-onnx-whisper-medium.en/resolve/main/test_wavs/trans.txt
|
||||
cd ../..
|
||||
mv $src ../..
|
||||
mv $src ../
|
||||
echo "pwd: $PWD"
|
||||
|
||||
cd ../..
|
||||
cd ../
|
||||
echo "--------------------"
|
||||
ls -lh
|
||||
ls -lh $src
|
||||
echo "--------------------"
|
||||
|
||||
tar cjvf ./$src.tar.bz2 $src
|
||||
if [[ $model == large || $model == large-v1 || $model == large-v2 || $model == distil-large-v2 ]]; then
|
||||
#tar cvjf - $src | split --bytes=1024MB - $src.tar.bz2.
|
||||
tar cvjf $src.tar.bz2 $src
|
||||
split -b 1G $src.tar.bz2 $src.tar.bz2.
|
||||
rm $src.tar.bz2
|
||||
# cat $src.tar.gz.* | tar xjf -
|
||||
else
|
||||
tar cvjf $src.tar.bz2 $src
|
||||
fi
|
||||
ls -lh
|
||||
|
||||
|
||||
- name: Release
|
||||
uses: svenstaro/upload-release-action@v2
|
||||
with:
|
||||
file_glob: true
|
||||
file: ./*.tar.bz2
|
||||
file: ./*.tar*
|
||||
overwrite: true
|
||||
repo_name: k2-fsa/sherpa-onnx
|
||||
repo_token: ${{ secrets.UPLOAD_GH_SHERPA_ONNX_TOKEN }}
|
||||
@@ -99,14 +121,21 @@ jobs:
|
||||
GIT_LFS_SKIP_SMUDGE=1 git clone https://huggingface.co/csukuangfj/sherpa-onnx-whisper-${{ matrix.model }} huggingface
|
||||
rm -rf huggingface/*
|
||||
|
||||
cp -av $src/* ./huggingface/
|
||||
if [[ $model == large || $model == large-v1 || $model == large-v2 || $model == distil-large-v2 ]]; then
|
||||
mv $src.tar* ./huggingface
|
||||
else
|
||||
cp -v $src/*.onnx ./huggingface
|
||||
cp -v $src/*tokens* ./huggingface
|
||||
cp -av $src/test_wavs ./huggingface
|
||||
fi
|
||||
|
||||
cd huggingface
|
||||
|
||||
git status
|
||||
ls -lh
|
||||
git lfs track "*.onnx"
|
||||
# git lfs track "*.ort"
|
||||
git lfs track "*gz*"
|
||||
git lfs track "*onnx*"
|
||||
|
||||
git add .
|
||||
git commit -m "upload ${{ matrix.model }}"
|
||||
git push https://csukuangfj:$HF_TOKEN@huggingface.co/csukuangfj/sherpa-onnx-whisper-${{ matrix.model }} main
|
||||
|
||||
@@ -90,7 +90,7 @@ jobs:
|
||||
./sherpa-onnx-zipformer-en-2023-06-26/test_wavs/8k.wav
|
||||
|
||||
- name: Start server for paraformer models
|
||||
if: matrix.model_type == 'paraformer'
|
||||
if: matrix.model_type == 'paraformer' && matrix.os != 'windows-latest'
|
||||
shell: bash
|
||||
run: |
|
||||
GIT_LFS_SKIP_SMUDGE=1 git clone https://huggingface.co/csukuangfj/sherpa-onnx-paraformer-bilingual-zh-en
|
||||
@@ -106,7 +106,7 @@ jobs:
|
||||
sleep 10
|
||||
|
||||
- name: Start client for paraformer models
|
||||
if: matrix.model_type == 'paraformer'
|
||||
if: matrix.model_type == 'paraformer' && matrix.os != 'windows-latest'
|
||||
shell: bash
|
||||
run: |
|
||||
python3 ./python-api-examples/offline-websocket-client-decode-files-paralell.py \
|
||||
|
||||
Reference in New Issue
Block a user