README.md

metadata

title: DriftCall
emoji: 🌀
colorFrom: indigo
colorTo: pink
sdk: static
pinned: true
license: apache-2.0
short_description: Voice concierge under schema drift — project site
tags:
  - rl
  - voice
  - indic
  - schema-drift
  - openenv
  - gemma-3n

DriftCall — project site

Static project site for DriftCall, the OpenEnv-compliant voice concierge env under mid-episode schema drift. Built with Vite + React + TypeScript, vendored @chenglou/pretext for hero text measurement, single saffron accent on dark editorial brutalism.

What this Space is

Pre-built static HTML/CSS/JS only — no server, no GPU, no model. The page embeds the live demo Space (DGXAI/driftcall-demo) in an iframe and links out to the env Space and the trained LoRA.

Sister artefacts

• 🤗 OpenEnv env Space — DGXAI/driftcall-env
• 🎙️ Voice demo Space — DGXAI/driftcall-demo
• 🧠 Trained LoRA — DGXAI/gemma-3n-e2b-driftcall-lora
• 📦 Source — github

Build

The site is regenerated from canonical sources at the repo root via deploy/frontend_space/build.sh:

cd DRIFTCALL
HF_TOKEN=hf_...  bash deploy/frontend_space/build.sh --push

The script runs npm ci && npm run build in frontend/, copies the produced dist/ into deploy/frontend_space/build/, drops this README on top, and hf uploads the result. The upload is just static files — no Docker, no Python, no runtime.