Maya Mathe Tutorial Video - Current Production Package

Current outcome

The initial production package is now grounded in the voice memo and Google Drive source folder.

Completed artifacts

1. Clean voice memo transcript

files/transcripts/maya_mathe_video_voice_memo_clean_transcript.md

1. Raw local transcript and timing segments

files/transcripts/maya_mathe_video_voice_memo_raw_transcript.txt

files/transcripts/maya_mathe_video_voice_memo_segments.json

1. Google Drive source index

files/maya_mathe_drive_source_index.md

1. Script/storyboard outline

files/maya_mathe_tutorial_video_script_storyboard_outline.md

1. Gemini creative prompt

files/creative/maya_mathe_gemini_creative_brief_prompt.md

1. Stitch visual prompt

files/creative/maya_mathe_stitch_visual_prompt.md

1. Creative sidekick status

files/creative/maya_mathe_creative_sidekick_status.md

1. Gemini creative direction

files/creative/maya_mathe_gemini_creative_direction.md

1. Stitch visual direction

files/creative/maya_mathe_stitch_visual_direction.md

1. Matias storyboard feedback integration

files/matias_storyboard_feedback_integrated.md

1. Catchy 10-second intro/outro concept

files/maya_mathe_intro_outro_concept.md

1. Raw 30-second first video draft

files/video_drafts/maya_mathe_raw_first_video_30s.mp4

1. ElevenLabs voice/music plan

files/elevenlabs_voice_music_plan.md

Storyline extracted from the memo

The video should be a school math-learning explainer using Maya and Emma's recorded voices as the main narration. The teacher cares especially about how the students explain the topic, so the voices must stay central.

The requested visual world is a blackboard/classroom scene. Math examples and key points should appear on the board as if written by a ghost hand. Two friendly object avatars can stand beside the board: Maya as a calculator and Emma as a ruler.

Matias feedback integrated

The storyboard scaffold has been corrected by matias_input_Storyboard_Outline_01.md.

Confirmed direction:

1. Maya is a real 15-year-old student, not a fictional presenter or generic brand.
2. The audience is other 10th-grade students and the math teacher.
3. Grade level is 10th grade, German Gymnasium.
4. Language is German, Hannover-Hochdeutsch.
5. Tone should fit a high-performing top-of-class 10th grader: eloquent, smart, and precise.
6. Visual identity should feel like a smart US-style school class, with 21 Jump Street as a mood reference, adapted safely for a school math explainer.
7. Runtime is confirmed at 5-8 minutes.
8. Master format is 16:9 with safe-center framing for 9:16 TikTok/Reels excerpts.
9. Final use is website embed plus TikTok/short-form excerpts.
10. First production priority is full video clips with animation, voiceover, and background music.
11. Second production priority is a management app for the video clips.
12. Maya still needs to be asked for fonts, colors, and intro/outro preferences.

Intro and outro concept

The video should include a catchy 10-second intro and a catchy 10-second outro.

Intro direction:

• quick chalkboard reveal
• Maya calculator avatar and Emma ruler avatar enter
• ghost hand writes Lineare Gleichungssysteme
• method labels appear: grafisch, Einsetzen, Addieren
• recommended tagline: Nicht nur rechnen: verstehen.

Outro direction:

• final solution box completes
• recap underline: Definition, 3 Verfahren, Sonderfaelle
• CTA: Arbeitsblatt herunterladen
• secondary CTA: Weiter zur naechsten Lektion
• animated finish flag waves beside the board

First raw video draft

A 30-second raw MP4 draft has been produced from the current briefing.

It uses:

• chalkboard scene
• Maya calculator avatar
• Emma ruler avatar
• ghost-writing style text
• original source narration from uploaded Drive audio

This is a showable proof-of-direction, not the final video.

ElevenLabs audio lane

MC confirmed that clear consent is already available for voice-avatar use.

ElevenLabs should be used for:

• Maya voice avatar
• Emma voice avatar
• intro/outro or repair lines
• background music
• chalk/ruler/calculator sound effects
• final audio mix support

The original uploaded student audio remains the primary narration source unless a later production decision approves broader synthetic replacement.

Source sequence from Drive

The source folder contains 15 items:

• 10 .m4a narration clips
• 5 WhatsApp image references
• 0 nested folders

Inferred source sequence:

1. Definition / introduction
2. Graphical solving
3. Graphical solving example
4. Substitution method with example
5. Equalization method
6. Equalization method example
7. Addition/elimination method with example
8. Special cases / solution sets
9. Special cases continuation or short correction
10. Summary

Seedance 2.0 production note

Seedance 2.0 is not locally installable. It is API/cloud-only. The researched production posture is to treat Seedance as optional remote AI-video support, not as a hard local dependency. The safer core workflow is to produce board-writing animation, character/avatar loops, voice sync, and edit assembly locally, with Seedance only used for optional generated supporting clips if access/licensing is resolved.

Current blockers

1. The 10 source .m4a clips still need transcripts and timing maps.
2. The 5 WhatsApp images still need content inspection and mapping to scenes.
3. Final export resolution and delivery packaging are still open.
4. Maya needs to provide preferred fonts, colors, and intro/outro style.

Recommended next production steps

1. Transcribe all 10 source narration clips from the Drive folder.
2. Download/inspect the five image references through gws only and map them to scenes.
3. Build a scene map aligning audio, board writing, image references, avatar beats, and captions.
4. Use the Gemini and Stitch direction files as the creative reference for avatar, board, and thumbnail design.
5. Ask Maya for brand/style preferences before final visual lock.