Skip to content

perf(whisper): reuse the language-detect encoder pass (1.9x on auto) #206

perf(whisper): reuse the language-detect encoder pass (1.9x on auto)

perf(whisper): reuse the language-detect encoder pass (1.9x on auto) #206

Workflow file for this run

name: CI
on:
pull_request:
branches: [main]
push:
branches: [main]
jobs:
test:
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v5
- uses: actions/setup-go@v6
with:
go-version: '1.25'
- name: Run tests
run: make test
- name: Build
run: go build -o zee
integration:
runs-on: macos-latest
needs: test
steps:
- uses: actions/checkout@v5
with:
submodules: recursive
- uses: actions/setup-go@v6
with:
go-version: '1.25'
# Cache the on-device Parakeet ggufs (~940MB) so the local-model tests
# run without re-downloading every time. Bump the key when the models
# release changes (localmodel.Version).
- name: Cache local models
uses: actions/cache@v4
with:
path: models/parakeet/v1
key: parakeet-models-v1
- name: Download local models
run: make download-models
- name: Integration tests
env:
GROQ_API_KEY: ${{ secrets.GROQ_API_KEY }}
DEEPGRAM_API_KEY: ${{ secrets.DEEPGRAM_API_KEY }}
run: make test-integration