Skip to content

Commit

Permalink
Add WebAssembly for ASR (#604)
Browse files Browse the repository at this point in the history
  • Loading branch information
@csukuangfj
csukuangfj authored Feb 23, 2024
1 parent a2df353 commit 16ba7e2
Show file tree
Hide file tree
Showing 21 changed files with 1,464 additions and 16 deletions.
142 changes: 142 additions & 0 deletions .github/workflows/wasm-simd-hf-space-en-asr-zipformer.yaml
View file
Original file line number Diff line number Diff line change
@@ -0,0 +1,142 @@
name: wasm-simd-hf-space-en-asr-zipformer

on:
release:
types:
- published

workflow_dispatch:

concurrency:
group: wasm-simd-hf-space-en-asr-zipformer-${{ github.ref }}
cancel-in-progress: true

jobs:
wasm-simd-hf-space-en-asr-zipformer:
runs-on: ${{ matrix.os }}
strategy:
fail-fast: false
matrix:
os: [ubuntu-latest]

steps:
- uses: actions/checkout@v4
with:
fetch-depth: 0
- name: Install emsdk
uses: mymindstorm/setup-emsdk@v14

- name: View emsdk version
shell: bash
run: |
emcc -v
echo "--------------------"
emcc --check
- name: Download model files
shell: bash
run: |
cd wasm/asr/assets
ls -lh
echo "----------"
wget -q https://github.com/k2-fsa/sherpa-onnx/releases/download/asr-models/sherpa-onnx-streaming-zipformer-en-2023-06-21.tar.bz2
tar xvf sherpa-onnx-streaming-zipformer-en-2023-06-21.tar.bz2
rm sherpa-onnx-streaming-zipformer-en-2023-06-21.tar.bz2
mv sherpa-onnx-streaming-zipformer-en-2023-06-21/encoder-epoch-99-avg-1.int8.onnx encoder.onnx
mv sherpa-onnx-streaming-zipformer-en-2023-06-21/decoder-epoch-99-avg-1.onnx decoder.onnx
mv sherpa-onnx-streaming-zipformer-en-2023-06-21/joiner-epoch-99-avg-1.onnx joiner.onnx
mv sherpa-onnx-streaming-zipformer-en-2023-06-21/tokens.txt ./
rm -rf sherpa-onnx-streaming-zipformer-en-2023-06-21
ls -lh
- name: Build sherpa-onnx for WebAssembly (ASR)
shell: bash
run: |
./build-wasm-simd-asr.sh
- name: collect files
shell: bash
run: |
SHERPA_ONNX_VERSION=v$(grep "SHERPA_ONNX_VERSION" ./CMakeLists.txt | cut -d " " -f 2 | cut -d '"' -f 2)
dst=sherpa-onnx-wasm-simd-${SHERPA_ONNX_VERSION}-en-asr-zipformer
mv build-wasm-simd-asr/install/bin/wasm/asr $dst
ls -lh $dst
tar cjfv ${dst}.tar.bz2 ./${dst}
- name: Upload wasm files
uses: actions/upload-artifact@v4
with:
name: sherpa-onnx-wasm-simd-en-asr-zipformer
path: ./sherpa-onnx-wasm-simd-*.tar.bz2

- name: Publish to ModelScope
env:
MS_TOKEN: ${{ secrets.MODEL_SCOPE_GIT_TOKEN }}
uses: nick-fields/retry@v2
with:
max_attempts: 20
timeout_seconds: 200
shell: bash
command: |
SHERPA_ONNX_VERSION=v$(grep "SHERPA_ONNX_VERSION" ./CMakeLists.txt | cut -d " " -f 2 | cut -d '"' -f 2)
git config --global user.email "[email protected]"
git config --global user.name "Fangjun Kuang"
rm -rf ms
export GIT_LFS_SKIP_SMUDGE=1
git clone https://www.modelscope.cn/studios/k2-fsa/web-assembly-asr-sherpa-onnx-en.git ms
cd ms
git fetch
git pull
git merge -m "merge remote" --ff origin main
cp -v ../sherpa-onnx-wasm-simd-${SHERPA_ONNX_VERSION}-*/* .
git status
git lfs track "*.data"
git lfs track "*.wasm"
ls -lh
git add .
git commit -m "update model"
git push https://oauth2:${MS_TOKEN}@www.modelscope.cn/studios/k2-fsa/web-assembly-asr-sherpa-onnx-en.git
- name: Publish to huggingface
env:
HF_TOKEN: ${{ secrets.HF_TOKEN }}
uses: nick-fields/retry@v2
with:
max_attempts: 20
timeout_seconds: 200
shell: bash
command: |
SHERPA_ONNX_VERSION=v$(grep "SHERPA_ONNX_VERSION" ./CMakeLists.txt | cut -d " " -f 2 | cut -d '"' -f 2)
git config --global user.email "[email protected]"
git config --global user.name "Fangjun Kuang"
rm -rf huggingface
export GIT_LFS_SKIP_SMUDGE=1
git clone https://huggingface.co/spaces/k2-fsa/web-assembly-asr-sherpa-onnx-en huggingface
cd huggingface
git fetch
git pull
git merge -m "merge remote" --ff origin main
cp -v ../sherpa-onnx-wasm-simd-${SHERPA_ONNX_VERSION}-*/* .
git status
git lfs track "*.data"
git lfs track "*.wasm"
ls -lh
git add .
git commit -m "update model"
git push https://csukuangfj:[email protected]/spaces/k2-fsa/web-assembly-asr-sherpa-onnx-en main
149 changes: 149 additions & 0 deletions .github/workflows/wasm-simd-hf-space-zh-en-asr-paraformer.yaml
View file
Original file line number Diff line number Diff line change
@@ -0,0 +1,149 @@
name: wasm-simd-hf-space-zh-en-asr-paraformer

on:
release:
types:
- published

workflow_dispatch:

concurrency:
group: wasm-simd-hf-space-zh-en-asr-paraformer-${{ github.ref }}
cancel-in-progress: true

jobs:
wasm-simd-hf-space-zh-en-asr-paraformer:
runs-on: ${{ matrix.os }}
strategy:
fail-fast: false
matrix:
os: [ubuntu-latest]

steps:
- uses: actions/checkout@v4
with:
fetch-depth: 0
- name: Install emsdk
uses: mymindstorm/setup-emsdk@v14

- name: View emsdk version
shell: bash
run: |
emcc -v
echo "--------------------"
emcc --check
- name: Download model files
shell: bash
run: |
cd wasm/asr/assets
ls -lh
echo "----------"
wget https://github.com/k2-fsa/sherpa-onnx/releases/download/asr-models/sherpa-onnx-streaming-paraformer-bilingual-zh-en.tar.bz2
tar xvf sherpa-onnx-streaming-paraformer-bilingual-zh-en.tar.bz2
rm sherpa-onnx-streaming-paraformer-bilingual-zh-en.tar.bz2
mv sherpa-onnx-streaming-paraformer-bilingual-zh-en/encoder.int8.onnx encoder.onnx
mv sherpa-onnx-streaming-paraformer-bilingual-zh-en/decoder.int8.onnx decoder.onnx
mv sherpa-onnx-streaming-paraformer-bilingual-zh-en/tokens.txt ./
rm -rf sherpa-onnx-streaming-paraformer-bilingual-zh-en
ls -lh
cd ../
sed -i.bak s/"type = 0"/"type = 1"/g ./sherpa-onnx.js
sed -i.bak s/Zipformer/Paraformer/g ./index.html
git diff
- name: Build sherpa-onnx for WebAssembly (ASR)
shell: bash
run: |
./build-wasm-simd-asr.sh
- name: collect files
shell: bash
run: |
SHERPA_ONNX_VERSION=v$(grep "SHERPA_ONNX_VERSION" ./CMakeLists.txt | cut -d " " -f 2 | cut -d '"' -f 2)
dst=sherpa-onnx-wasm-simd-${SHERPA_ONNX_VERSION}-zh-en-asr-paraformer
mv build-wasm-simd-asr/install/bin/wasm/asr $dst
ls -lh $dst
tar cjfv ${dst}.tar.bz2 ./${dst}
- name: Upload wasm files
uses: actions/upload-artifact@v4
with:
name: sherpa-onnx-wasm-simd-zh-en-asr-paraformer
path: ./sherpa-onnx-wasm-simd-*.tar.bz2

- name: Publish to ModelScope
env:
MS_TOKEN: ${{ secrets.MODEL_SCOPE_GIT_TOKEN }}
uses: nick-fields/retry@v2
with:
max_attempts: 20
timeout_seconds: 200
shell: bash
command: |
SHERPA_ONNX_VERSION=v$(grep "SHERPA_ONNX_VERSION" ./CMakeLists.txt | cut -d " " -f 2 | cut -d '"' -f 2)
git config --global user.email "[email protected]"
git config --global user.name "Fangjun Kuang"
rm -rf ms
export GIT_LFS_SKIP_SMUDGE=1
git clone https://www.modelscope.cn/studios/k2-fsa/web-assembly-asr-sherpa-onnx-zh-en-paraformer.git ms
cd ms
git fetch
git pull
git merge -m "merge remote" --ff origin main
cp -v ../sherpa-onnx-wasm-simd-${SHERPA_ONNX_VERSION}-*/* .
git status
git lfs track "*.data"
git lfs track "*.wasm"
ls -lh
git add .
git commit -m "update model"
git push https://oauth2:${MS_TOKEN}@www.modelscope.cn/studios/k2-fsa/web-assembly-asr-sherpa-onnx-zh-en-paraformer.git
- name: Publish to huggingface
env:
HF_TOKEN: ${{ secrets.HF_TOKEN }}
uses: nick-fields/retry@v2
with:
max_attempts: 20
timeout_seconds: 200
shell: bash
command: |
SHERPA_ONNX_VERSION=v$(grep "SHERPA_ONNX_VERSION" ./CMakeLists.txt | cut -d " " -f 2 | cut -d '"' -f 2)
git config --global user.email "[email protected]"
git config --global user.name "Fangjun Kuang"
rm -rf huggingface
export GIT_LFS_SKIP_SMUDGE=1
git clone https://huggingface.co/spaces/k2-fsa/web-assembly-asr-sherpa-onnx-zh-en-paraformer huggingface
cd huggingface
git fetch
git pull
git merge -m "merge remote" --ff origin main
cp -v ../sherpa-onnx-wasm-simd-${SHERPA_ONNX_VERSION}-*/* .
git status
git lfs track "*.data"
git lfs track "*.wasm"
ls -lh
git add .
git commit -m "update model"
git push https://csukuangfj:[email protected]/spaces/k2-fsa/web-assembly-asr-sherpa-onnx-zh-en-paraformer main
Loading

0 comments on commit 16ba7e2

Please sign in to comment.