name: export-sense-voice-to-onnx

on:
  workflow_dispatch:

concurrency:
  group: export-sense-voice-to-onnx-${{ github.ref }}
  cancel-in-progress: true

jobs:
  export-sense-voice-to-onnx:
    if: github.repository_owner == 'k2-fsa' || github.repository_owner == 'csukuangfj'
    name: export sense-voice
    runs-on: ${{ matrix.os }}
    strategy:
      fail-fast: false
      matrix:
        os: [ubuntu-latest]
        python-version: ["3.10"]

    steps:
      - uses: actions/checkout@v4

      - name: Setup Python ${{ matrix.python-version }}
        uses: actions/setup-python@v5
        with:
          python-version: ${{ matrix.python-version }}

      - name: Download test_wavs
        shell: bash
        run: |
          sudo apt-get install -y -qq sox libsox-fmt-mp3
          curl -SL -O https://huggingface.co/FunAudioLLM/SenseVoiceSmall/resolve/main/example/zh.mp3
          curl -SL -O https://huggingface.co/FunAudioLLM/SenseVoiceSmall/resolve/main/example/en.mp3
          curl -SL -O https://huggingface.co/FunAudioLLM/SenseVoiceSmall/resolve/main/example/ja.mp3
          curl -SL -O https://huggingface.co/FunAudioLLM/SenseVoiceSmall/resolve/main/example/ko.mp3
          curl -SL -O https://huggingface.co/FunAudioLLM/SenseVoiceSmall/resolve/main/example/yue.mp3

          soxi *.mp3

          sox zh.mp3 -r 16k zh.wav
          sox en.mp3 -r 16k en.wav
          sox ja.mp3 -r 16k ja.wav
          sox ko.mp3 -r 16k ko.wav
          sox yue.mp3 -r 16k yue.wav

      - name: Run
        shell: bash
        run: |
          cd scripts/sense-voice
          ./run.sh

      - name: Publish to huggingface
        env:
          HF_TOKEN: ${{ secrets.HF_TOKEN }}
        uses: nick-fields/retry@v3
        with:
          max_attempts: 20
          timeout_seconds: 200
          shell: bash
          command: |
            git config --global user.email "csukuangfj@gmail.com"
            git config --global user.name "Fangjun Kuang"

            rm -rf huggingface
            export GIT_LFS_SKIP_SMUDGE=1
            export GIT_CLONE_PROTECTION_ACTIVE=false

            git clone https://csukuangfj:$HF_TOKEN@huggingface.co/csukuangfj/sherpa-onnx-sense-voice-zh-en-ja-ko-yue-2024-07-17 huggingface
            cd huggingface
            git fetch
            git pull
            echo "pwd: $PWD"
            ls -lh ../scripts/sense-voice

            rm -rf ./*

            cp -v ../scripts/sense-voice/*.onnx .
            cp -v ../scripts/sense-voice/tokens.txt .
            cp -v ../scripts/sense-voice/README.md .
            cp -v ../scripts/sense-voice/export-onnx.py .

            mkdir test_wavs
            cp -v ../*.wav ./test_wavs/

            curl -SL -O https://raw.githubusercontent.com/FunAudioLLM/SenseVoice/main/LICENSE

            git lfs track "*.onnx"
            git add .

            ls -lh

            git status

            git commit -m "add models"
            git push https://csukuangfj:$HF_TOKEN@huggingface.co/csukuangfj/sherpa-onnx-sense-voice-zh-en-ja-ko-yue-2024-07-17 main || true

            cd ..

            rm -rf huggingface/.git*
            dst=sherpa-onnx-sense-voice-zh-en-ja-ko-yue-2024-07-17

            mv huggingface $dst

            tar cjvf $dst.tar.bz2 $dst
            rm -rf $dst

      - name: Release
        uses: svenstaro/upload-release-action@v2
        with:
          file_glob: true
          file: ./*.tar.bz2
          overwrite: true
          repo_name: k2-fsa/sherpa-onnx
          repo_token: ${{ secrets.UPLOAD_GH_SHERPA_ONNX_TOKEN }}
          tag: asr-models