mirror of
https://github.com/hanzoai/video.git
synced 2026-10-30 18:18:27 +00:00
calesthio
355fd2b7d6
Implements the 2026-04-02 transformation spec (Phases 1-8) and fixes all critical bugs found during 5-pipeline E2E testing. Governance & Decision Intelligence: - Pipeline-specific stage order in checkpoint (replaces global STAGES list) - Provider scoring engine (lib/scoring.py) with 7-dimension weighted ranking - Decision log artifact enforced at proposal/idea stage across all 10 pipelines - Delivery promise classifier prevents silent motion-to-still downgrades - Structured shot language in scene_plan schema (camera, lens, lighting, DOF) - Variation checker and slideshow risk scorer block samey output before render - Creative intake, capability extension, and creative-intake meta skills - Final self-review artifact with 5 mandatory checks before presenting output - Source media review contract for user-supplied footage Render & Theme System: - Remotion AnimatedBackground now derives colors from playbook (no more hardcoded dark blue fintech gradient on every video) - video_compose builds custom ThemeConfig from playbook YAML colors/fonts — custom playbooks flow through to Remotion automatically - Explainer component wires theme to all child components (charts, cards, etc.) - resolveAsset() handles absolute paths on Windows/Unix via file:// URIs - RENDERER_FAMILY_MAP synced with actual Remotion compositions Critical Bug Fixes: - Windows npx subprocess: run_command() resolves .cmd wrappers via shutil.which() - Silent renderer downgrade: Remotion failure now returns explicit error with options instead of silently falling back to FFmpeg - .env inline comment parsing strips trailing # comments from API keys - concat_path UnboundLocalError in video_compose finally block - audio_mixer and showcase_card capture=True kwarg bug - Selector estimate_cost() calls fixed (_select_tool -> _select_best_tool) - asset_manifest schema expanded with provider, license, subtype fields - screen-demo subtitle_gen moved from required to optional tools - Duration drift detection in post-render final review (>25% warns)
2.3 KiB
2.3 KiB
Script Director — Talking Head Pipeline
When to Use
You have a brief and raw talking-head footage. Your job is to transcribe the footage and structure it into a script artifact with timestamped sections.
Unlike the explainer pipeline (which writes a script from scratch), you're extracting and structuring existing speech.
Prerequisites
| Layer | Resource | Purpose |
|---|---|---|
| Schema | schemas/artifacts/script.schema.json |
Artifact validation |
| Prior artifacts | state.artifacts["idea"]["brief"] |
Content context |
| Tools | transcriber (WhisperX) |
Speech-to-text with timestamps |
Process
Step 1: Transcribe
Use the transcriber tool to get word-level timestamps:
- Model:
large-v3for best quality,basefor speed - Enable word-level alignment for precise timing
- Note language detection result
Step 2: Segment into Sections
Group the transcript into logical sections:
- Detect topic changes by content
- Respect natural pauses (> 1.5s silence = potential section break)
- Each section gets: id, text, start_seconds, end_seconds
Step 3: Enhance Section Metadata
For each section, add:
- Enhancement cues (where overlays, b-roll, or text cards could go)
- Speaker notes (emphasis, pace changes detected in audio)
Step 4: Build Script Artifact
Assemble the structured script with:
- Total duration (from transcript)
- All sections with timestamps
- Enhancement cues per section
Step 5: Self-Evaluate
| Criterion | Question |
|---|---|
| Transcription accuracy | Are the words correct? (Spot-check a few sections) |
| Timestamp accuracy | Do section boundaries align with actual speech? |
| Coverage | Does the script span the full footage duration? |
Step 6: Submit
Validate the script against the schema and persist via checkpoint.
Mid-Production Fact Verification
If you encounter uncertainty during script writing:
- Use
web_searchto verify factual claims before committing them to the script - Use
web_searchto find reference images for visual accuracy - Log verification in the decision log:
category="visual_accuracy_check"
Every factual claim in the script should be traceable to the research_brief.
If you make a claim that isn't in the research, do additional research and
add the source. Do not invent statistics, dates, or attributions.