Visual Pipeline

Status: Approved initial pipeline

Production flow

1. Approve the episode concept, learning goal, and Suno story/song brief.
2. Generate the episode’s music, two-character voices, spoken narration, spoken-word tags, and emotional performance through Suno.
3. Select and lock one finished Suno version.
4. Import that exact song into Palmier as the authoritative episode audio.
5. Run Whisper against the imported song to produce a timecoded transcript.
6. Review the Whisper transcript against the approved lyrics and narration, correct recognition errors, and preserve verified timecodes.
7. Convert the reviewed transcript into the shot timing map.
8. Lock the episode’s cast, guest, location, and style records.
9. Write a 12-panel board against the Whisper/Palmier timing using the coverage rules in Little Lanterns Format.
10. Assign every panel a transcript cue and exact Palmier audio range.
11. Compile each structured image record through the Anima model adapter.
12. Generate Anima storyboard candidates locally in ComfyUI.
13. Review character recognition, location continuity, shot diversity, story clarity, and the motion opportunity in every frame.
14. Approve one Anima frame per shot. That image becomes the first frame and visual continuity contract for image-to-video.
15. Write a motion prompt that describes only what changes after the approved first frame during its assigned audio range.
16. Animate each approved frame through the local Wan 2.2 14B image-to-video workflow.
17. Return the selected Wan video shots to the existing Palmier project.
18. Assemble, time, revise, and finalize all episode media in Palmier.
19. Store review assets in SmartGallery and record the Palmier project, timeline, final outputs, links, and provenance here.

Handoff contracts

Suno → Palmier → Whisper timing

Suno is always the first generated-media stage. Every locked audio version hands off:

audio_handoff:
  episode: S01E__
  suno_prompt_record: ""
  music_output: ""
  narration_output: ""
  selected_version: ""
  duration: ""
  palmier_project: ""
  palmier_audio_asset: ""
  palmier_audio_track: ""
  whisper_model: ""
  whisper_workflow_version: ""
  whisper_transcript: ""
  transcript_review_status: Draft
  emotional_arc: ""
  spoken_word_tags: []
  timing_map:
    - cue_id: CUE-01
      start: ""
      end: ""
      lyric_or_narration: ""
      story_beat: ""
      visual_opportunity: ""
      transcript_source: whisper
  locked_on: YYYY-MM-DD

Music and narration may be generated as one integrated performance or as separate approved outputs. In either case, the exact finished audio used in Palmier is the file Whisper must transcribe. The transcript text is reviewed against the approved lyrics and narration before its timecodes become the visual-production authority.

Timed visual plan and Anima → Wan 2.2 14B

Every approved storyboard shot hands off:

shot_handoff:
  shot_id: S01E__-SH__
  approved_first_frame: ""
  anima_prompt_record: ""
  anima_workflow_version: ""
  seed: ""
  dimensions: 864x1152
  visible_characters: []
  continuity_anchors: []
  whisper_cue_ids: []
  audio_start: ""
  audio_end: ""
  intended_action: ""
  camera_motion: ""
  environmental_motion: ""
  wan_motion_prompt: ""
  wan_workflow_version: ""

The first-frame image owns appearance, framing, wardrobe, location, props, and starting pose. The Wan motion prompt owns change over time. It must not redesign the frame.

Wan and Suno → Palmier

Palmier is the approved finishing environment for all media for now. Each episode hands off:

palmier_handoff:
  episode: S01E__
  palmier_project: ""
  timeline_version: ""
  selected_wan_shots: []
  selected_suno_music: ""
  selected_suno_narration: ""
  edit_order: []
  timing_notes: ""
  revisions: []
  final_output: ""
  final_status: Draft

Palmier owns assembly, editorial timing, shot trimming, audio placement, revision tracking, and the finalized media output. It does not replace the source-generation records for Anima, Wan, or Suno.

Current image baseline

Field	Candidate
Orientation	Portrait video
Aspect ratio	3:4
Working dimensions	864 × 1152
Board size	12 panels
Storyboard / first-frame model	Anima
Image-to-video model	Wan 2.2 14B
Music and narration	Suno
Finishing environment	Palmier (approved interim destination)
Character strategy	Distinctive visible anchors + structured prompt inheritance
Review strategy	Contact sheet first, individual frames second

Visual acceptance checklist

• Mia and Henry are recognizable without relying on labels.
• Close-ups are genuinely close; the face occupies most of the frame.
• The sequence includes wide, medium, close-up, ECU, two-shot, OTS, and reverse coverage.
• Camera position, subject scale, and composition meaningfully change between panels.
• Style phrasing remains unchanged across the sequence.
• Location anchors recur without forcing identical backgrounds.
• Hands, gaze, props, and screen direction support the intended edit.
• No unexplained duplicate character or wardrobe mutation appears.
• Every approved frame has a specific, achievable motion plan.
• Wan preserves the approved first-frame identity, composition, and location.
• Every panel and Wan shot references its reviewed Whisper cue and exact Palmier audio range.
• Music and narration establish the emotional beats interpreted by the storyboard.
• The transcript was reviewed against the approved lyrics and narration.
• The Whisper timing source is the exact Suno file imported into Palmier.
• The Palmier timeline references the selected Wan and Suno versions.
• The final Palmier output can be traced back to every source-generation record.

Known lesson

Model quality alone does not produce a storyboard. Shot type must be expressed as camera geometry and subject scale, and each panel must have a distinct editorial job.