Overview

Appendix A: Google GenAI Media Tooling & Educational Asset Catalog

Playbook: PB-02 (AI Video Making for Kids Educational Media)
Tooling Stack: Google Gemini 2.5 Flash, Imagen 3, Veo 2, Cloud TTS Journey, DeepMind Lyria, FFmpeg 7.0+
Purpose: Production Sizing, Quick-Reference API Parameter Cheat Sheet & Master Asset Catalog
Target Audience: Year 1 Computer Science & Software Engineering Students

1. Quick-Reference Google GenAI API Matrix

2. Google Imagen 3 Production Parameter Cheat Sheet

Model Specification

• Model ID: imagen-3.0-generate-002
• Recommended Aspect Ratio: "16:9" for video keyframes, "1:1" for orthographic turnarounds.
• Color Grading: Standard Rec.709 sRGB gamut.

Master Stylistic Prompt Prefixes (Copy-Paste)

3D Pixar Animation Style (Recommended for Toddler Media)

[3D Stylized Pixar Animation Style, clean matte clay and soft felt finish, warm soft studio key light, gentle rim lighting, zero harsh cast shadows, pastel educational studio backdrop]:

Soft Storybook Claymation Style

[Stop-motion claymation style, handcrafted plasticine textures, soft tactile thumbprint details, warm tungsten miniature lighting, pastel storybook diorama]:

Universal Negative Prompt Suite for Kids Media

photorealistic, realistic human skin, human pores, sharp teeth, fangs, creepy, terrifying, uncanny valley, realistic animal fur, realistic feathers, strabismus, asymmetrical eyes, wall-eyed, extra limbs, extra fingers, missing limbs, mutated hands, claw deformities, background clutter, noisy background, high contrast shadows, dark atmosphere, text, watermark, signature

3. Google Veo 2 (`veo-2.0`) Directing Cheat Sheet

Kinematic Constraints for Toddlers

• Camera Pan Velocity: Maximum $0.05\text{ m/s}$ (gentle micro-pans only).
• Camera Roll: Strictly $0.0^\circ$.
• Cognitive Stillness: At least $3.0\text{ seconds}$ of stable held framing per vocabulary shot.
• Frame Rate: Strictly $24\text{ fps}$ (constant frame rate).

Approved Camera Motion Directives

# 1. Locked Tripod (Vocabulary introduction & phonics)
Camera Setup: Locked static tripod, eye-level medium shot, zero camera panning, zero tilt, zero roll.

# 2. Subtle Dolly-In (Call-and-response listening beat)
Camera Setup: Subtle smooth 5% dolly-in toward subject at 0.02 m/s, eye-level framing, zero roll.

# 3. Gentle Low-Angle Pedestal (Celebration hop)
Camera Setup: Gentle upward pedestal move of 10cm at 0.03 m/s, maintaining subject centered in frame.

# 4. Slow Dolly-Out (Outro goodbye wave)
Camera Setup: Smooth slow dolly-out of 15% at 0.03 m/s, revealing wide cheerful stage setting.

Negative Temporal Prompt Suite for Veo 2

camera shake, fast camera pan, camera rotation, camera roll, whip pan, handheld jitter, limb morphing, melting geometry, rubbery limbs, extra limbs, disappearing costume, flickering textures, noise shimmer, frame distortion, character floating, teleportation, duplicate characters, text, watermark

4. Google Cloud TTS Journey SSML Reference

Recommended Voice IDs

• en-US-Journey-F: Warm, empathetic female voice (ideal for toddler mascot teacher).
• en-US-Journey-O: Friendly, gentle male voice.

Master SSML Prosody Templates

Template 1: Encounter Beat (Speed: 85%, Pitch: +12%)

<speak>
  <prosody rate="85%" pitch="+12%">
    Look! <break time="300ms"/>
    It is a <emphasis level="strong">${WORD}</emphasis>!
  </prosody>
</speak>

Template 2: Phonics Beat (Speed: 75%, Pitch: +15%)

<speak>
  <prosody rate="75%" pitch="+15%">
    ${LETTER_1} <break time="200ms"/> 
    ${LETTER_2} <break time="200ms"/> 
    ${LETTER_3}. <break time="300ms"/> 
    <emphasis level="strong">${WORD}</emphasis>!
  </prosody>
</speak>

Template 3: Call-and-Response Beat (With 2,500ms Child Pause)

<speak>
  <prosody rate="85%" pitch="+15%">
    Say <emphasis level="strong">${WORD}</emphasis>!
  </prosody>
  <break time="2500ms"/>
  <prosody rate="90%" pitch="+20%">
    Hooray!
  </prosody>
</speak>

5. Google DeepMind Lyria / MusicFX Prompt Recipes

Recipe 1: Standard Preschool Nursery Rhyme Bed (108 BPM)

[Style: Preschool Educational Nursery Bed, Tempo: 108 BPM, Key: C Major, Mood: Sunny, Gentle, Welcoming, Safe]: 
Instrumentation: Bright wooden marimba, playful pizzicato violins, light acoustic ukulele strumming, cheerful glockenspiel top melody, soft wooden egg shaker. 
Structure: Simple repetitive 4-bar nursery motif, transparent acoustic mix, soft low-end, zero distorted synths, zero minor chords, master educational bed.

Recipe 2: Upbeat Celebration Outro Dance (112 BPM)

[Style: Toddler Celebration Dance, Tempo: 112 BPM, Key: G Major, Mood: Joyful, Triumphant, Playful]: 
Instrumentation: Upbeat xylophone, bouncy acoustic bass, clapping rhythm, cheerful brass stabs (soft), acoustic guitar strumming, celebratory glockenspiel. 
Structure: Energetic celebratory 8-bar chorus, festive nursery finale, uniform volume.

6. Advanced SubStation Alpha (`.ass`) Syllable Karaoke Cheat Sheet

Style Definition Header

[Script Info]
Title: Kids Educational Bouncy Karaoke Subtitles
ScriptType: v4.00+
PlayResX: 1920
PlayResY: 1080

[V4+ Styles]
Format: Name, Fontname, Fontsize, PrimaryColour, SecondaryColour, OutlineColour, BackColour, Bold, Italic, Underline, StrikeOut, ScaleX, ScaleY, Spacing, Angle, BorderStyle, Outline, Shadow, Alignment, MarginL, MarginR, MarginV, Encoding
Style: KidsKaraoke,Arial Rounded MT Bold,64,&H00FFFFFF,&H0033FFFF,&H004C2B1A,&H80000000,-1,0,0,0,100,100,2,0,1,4.5,2.0,2,80,80,90,1

[Events]
Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text

Syllable Highlight Tag Syntax

• {\k<centiseconds>}: Progressively fills syllable with SecondaryColour (Canary Yellow &H0033FFFF&).
• Example: Dialogue: 0,0:00:10.00,0:00:13.50,KidsKaraoke,,0,0,0,,{\k25}Look! {\k20}It {\k20}is {\k20}a {\k50\b1}COW!{\b0}

7. FFmpeg 7.0+ Master Post-Production Snippets

Snippet 1: Lossless Video Shot Concatenation

# Create shots list
cat <<EOF > shots.txt
file 'shot_01.mp4'
file 'shot_02.mp4'
file 'shot_03.mp4'
file 'shot_04.mp4'
file 'shot_05.mp4'
file 'shot_06.mp4'
file 'shot_07.mp4'
file 'shot_08.mp4'
file 'shot_09.mp4'
file 'shot_10.mp4'
file 'shot_11.mp4'
EOF

# Concat video
ffmpeg -y -f concat -safe 0 -i shots.txt -c copy assembled_video.mp4

Snippet 2: Dynamic Sidechain Audio Ducking (-12dB)

ffmpeg -y \
  -i bg_music.wav \
  -i dialogue.wav \
  -filter_complex "\
    [1:a]asplit=2[dia_out][dia_sc]; \
    [0:a][dia_sc]sidechaincompress=threshold=0.035:ratio=6:attack=50:release=350:makeup=1.0[mus_ducked]; \
    [mus_ducked]volume=0.35[mus_bal]; \
    [mus_bal][dia_out]amix=inputs=2:duration=longest:weights=1.0 1.0[outa]" \
  -map "[outa]" -ar 48000 -c:a pcm_s16le master_soundtrack.wav

Snippet 3: Burn Bouncy Subtitles & Final YouTube Kids Export

ffmpeg -y \
  -i assembled_video.mp4 \
  -i master_soundtrack.wav \
  -vf "ass=karaoke_subtitles.ass" \
  -c:v libx264 -preset medium -crf 18 -pix_fmt yuv420p \
  -c:a aac -b:a 192k -ar 48000 \
  -shortest final_episode_1080p.mp4

8. CEFR Pre-A1 & A1 Early Childhood Theme Catalog

9. Production Sizing & Infrastructure Economics

Per-Episode Resource Consumption (60.0 Seconds, 11 Shots)

• Total Gemini API Calls: 1 call (~1,200 prompt tokens, ~2,500 output tokens) $\rightarrow$ $0.005
• Total Imagen 3 Keyframe Generations: 12 images (1 turnaround + 11 shot keys) $\rightarrow$ $0.48
• Total Google Veo 2 Generations: 11 clips @ 5s $\rightarrow$ $2.42
• Total Cloud TTS Generations: 11 utterances (~450 characters) $\rightarrow$ $0.015
• Total Lyria Music Generation: 1 track (60s) $\rightarrow$ $0.10
• FFmpeg Assembly Compute: ~15 seconds on 4-core CPU $\rightarrow$ $0.00
• Total Marginal Cost per 60-Second Episode: ~$3.02

Scaled Production Economics

• Daily Capacity (Automated VPS): 50 episodes / day.
• Monthly Output: 1,500 episodes / month.
• Total Monthly Cloud Cost: ~$4,530 (yielding 25 hours of broadcast-quality 1080p educational content).