// Run tests. Use after code changes to validate. Arguments: unit (default, no GPU), e2e (with models), filter name, or all.
Run performance benchmarks. Arguments: asr, tts, vad, diarize (with optional --num-files N).
Build the speech-swift package (release or debug). Use when preparing for testing, benchmarking, or running demos.
| name | test |
| description | Run tests. Use after code changes to validate. Arguments: unit (default, no GPU), e2e (with models), filter name, or all. |
Run unit and/or E2E tests.
unit (default): Quick tests, no model downloads. CI-safe.e2e: Full pipeline tests with real models. Requires metallib + cached weights.all: Both unit and E2E.--filter to swift test.arg="${ARGUMENTS:-unit}"
case "$arg" in
unit) swift test --skip E2E 2>&1 | tail -20 ;;
e2e) swift test 2>&1 | tail -30 ;;
all) swift test 2>&1 | tail -30 ;;
*) swift test --filter "$arg" 2>&1 | tail -20 ;;
esac
E2E test classes MUST be prefixed with E2E. Unit test classes must NOT contain E2E.