Inject hf_hub_download directly into diac module to work around import failures

Patch hf_hub_download to translate use_auth_token to token for newer huggingface_hub

Pre-download Diac model into image cache for HF Spaces

Fix .dockerignore to include frontend source for HF Spaces build

Fix short_description length for HF Spaces metadata

Merge HF Space initial commit, keep our README

Adapt project for Hugging Face Spaces deployment

initial commit

Add production-ready Docker deployment for local server hosting

Remove unwanted character from App.tsx

last

Fallback .env parser when python-dotenv is missing

Improve Supabase env diagnostics in /health and prefer .env values

Load .env automatically for Supabase and backend settings

Add editable transcription workflow and Supabase save

Fix mic upload: record WAV; light Arabic UI without emojis

Redesign frontend as Arabic-only شَمْل voice demo

Fix Whisper short-audio error: pad clips to 30s mel window

Add audio upload: Whisper ASR + diacritization pipeline

Add GET / root endpoint with API links for easier local verification

Use text branch for text-only inference on text+ASR models

Fix text-only inference for text+ASR Diac models

Register legacy tokenizer module alias for Diac checkpoints

Fix Lightning patch: patch pl_load instead of restricted classmethod

Document PyTorch 2.6 checkpoint load fix in README

Fix PyTorch 2.6 checkpoint loading for Diac Lightning models

Ignore frontend build artifacts (tsbuildinfo)

Add Diac API backend, Arabic RTL frontend, Docker, and docs

add files

add files