Skip to content

feat: add vsl_no sign pack [FaaFyfxR9WAQrL7FcAgEHJvztd8cVMxvjHRS55rw1nwH] - #574

Open
waterWang wants to merge 2 commits into
mergeos-bounties:masterfrom
waterWang:feat/vsl-no
Open

feat: add vsl_no sign pack [FaaFyfxR9WAQrL7FcAgEHJvztd8cVMxvjHRS55rw1nwH]#574
waterWang wants to merge 2 commits into
mergeos-bounties:masterfrom
waterWang:feat/vsl-no

Conversation

@waterWang

Copy link
Copy Markdown

Summary

Add isolated sign vsl_no (VSL negation) to Loru so sign to text / sign to voice can learn it.

Changes

  • data/samples/vsl_no.json — 8-frame, 21-landmark synthetic prototype
  • src/loru/models/vocab.py — register vsl_no in DEFAULT_GLOSS
  • src/loru/infer/text.py — sentence template vsl_no to No.
  • tests/test_vsl_no.py — 4 unit tests

Verification

pytest tests/test_vsl_no.py -v # 4 passed
pytest tests/ -v # 38 passed
ruff check src tests # all passed

Fixes #179

Wallet: FaaFyfxR9WAQrL7FcAgEHJvztd8cVMxvjHRS55rw1nwH

wangye and others added 2 commits July 25, 2026 17:55
…geos-bounties#10)

Adds:
- POST /infer/voice endpoint accepting sequence frames → returns audio path
- Test coverage for voice endpoint (valid request, missing frames)
- Test coverage for /infer/text missing frames edge case
- OpenAPI tags match existing endpoints

Testing: 5 tests pass with TestClient
  ✓ test_health
  ✓ test_sentence
  ✓ test_voice
  ✓ test_voice_no_frames
  ✓ test_infer_text_no_frames

[FaaFyfxR9WAQrL7FcAgEHJvztd8cVMxvjHRS55rw1nwH]
…[FaaFyfxR9WAQrL7FcAgEHJvztd8cVMxvjHRS55rw1nwH]
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

[25 MRG] Sign pack: vsl_no (vsl) — photo/video + landmarks

1 participant