- Voice clone integration: When user selects voice clone in Write phase,
backend uses their uploaded voice sample + scene script text to generate
audio via qwen3/minimax/cosyvoice voice clone APIs
- Multi-tenant workspace storage: All podcast assets (audio, video, images,
charts) now use workspace-specific directories per user
- Chart preview improvements: Card-based B-Roll charts UI with thumbnails,
takeaway text, and action buttons; public endpoint for image serving
- Voice clone caching: In-memory LRU cache for voice samples (avoids
re-downloading per scene); frontend caches voice clone metadata
- Thread pool for voice clone: Audio generation uses ThreadPoolExecutor to
avoid blocking the FastAPI event loop
- Auto-detect voice clone IDs (vc_*, MY_VOICE_CLONE) to route correctly
- DB fallback for voice sample URL: Fetches from ContentAsset if not passed
- Fixed API URL resolution for chart previews
- Fixed GlassyCard DOM warnings for motion props
- Fixed ScriptGenerationProgressView syntax error
- Fixed usePodcastWorkflow scriptData reference
The save_file_safely function may add a UUID suffix to the filename if there's
a collision, but the preview URL was being constructed using the original
filename instead of the actual saved filename. Now uses saved_preview_path.name.
- Remove nltk dependency from step4_assets by inlining _extract_user_id
- Add graceful error handling for step4_assets in podcast-only mode
- Fix media_utils.py to use tenant-aware get_podcast_media_read_dirs()
- Resolves avatar image 404 during scene image generation
- Fix database session handling in main_image_editing.py to use proper generator handling
- Add graceful handling of validation errors in podcast-only mode
- Add better error messages when WAVESPEED_API_KEY or HF_TOKEN is missing
- Add specific HTTP 503 error for configuration issues
- Add ALWRITY_SKIP_IMAGE_EDITING_VALIDATION env var to bypass validation in dev
In podcast-only demo mode, core routers were being skipped entirely,
which caused the voice clone endpoint (/onboarding/assets/create-voice-clone)
to return 404. Now podcast-enabled routers from CORE_ROUTER_REGISTRY
are included even in podcast-only mode.
- Add pandas to requirements-podcast.txt for usage tracking
- Fix LLM prompt to return plain strings instead of objects for enhanced_ideas
- Add object-to-string normalization for LLM responses that return objects
- Skip bible generation in podcast-only mode (onboarding disabled)
- Skip alerts polling in AlertsBadge when in podcast-only demo mode
- Conditionally import component_logic_router only when NOT in podcast mode
- Conditionally import seo_tools_router only when NOT in podcast mode
- Both use seo_analyzer which requires beautifulsoup4
- Also added debug logging to render-build.sh to verify ALWRITY_ENABLED_FEATURES
- Added beautifulsoup4 to requirements-podcast.txt (was missing)
