Batch Production Template �?" 15-Video Batch Tracker and Release Gate
Batch ID: _____________ Book: __________ Author: __________ Editor: __________ Date started: __________ Date completed: __________ Campaign ID: __________ Source brief: _____________
Batch Standard
This template tracks one standard production unit: one batch of 15 short videos.
- Each video has one preferred 10-second scene; two scenes are the absolute maximum.
- Each video uses one approved voiceover that must occupy the same 10-second scene.
- The complete production brief is assembled in a Google Doc before any video is generated.
- TikTok Symphony generates vertical B-roll from scene-specific prompts.
- Google Vids imports the document; human review and editing are mandatory.
- No single default Google Vids voice narrates all 15 videos.
- Weekly TikTok Symphony planning assumption: 400 credits; initial generation 15 x 10 = 150 credits; reserved correction allowance 250 credits.
- Verify the actual balance, allowance, and per-generation cost before starting. Do not treat planning assumptions as platform guarantees.
Credit Budget Tracker
| Item | Value |
|---|---|
| Opening balance (verified) | |
| Initial generation allocation (15 x 10) | 150 |
| Actual initial generation expenditure | |
| Correction / regeneration expenditure | |
| Remaining balance | |
| Outstanding corrections |
Rule: Never exceed the verified available balance. Fix text, narration, timing, and assembly errors in the editor when possible. Regenerate only when the visual is unusable or factually wrong.
Voice Allocation Plan (suggested)
Use five voice groups with three videos per group where the available voices and book content support it. Reuse voices intelligently, but do not use one voice for all 15 videos.
| Voice group | Assigned voice | Video IDs | Audience / tone fit | Pronunciation notes |
|---|---|---|---|---|
| Group 1 | V01, V02, V03 | |||
| Group 2 | V04, V05, V06 | |||
| Group 3 | V07, V08, V09 | |||
| Group 4 | V10, V11, V12 | |||
| Group 5 | V13, V14, V15 |
If voice availability is limited, document the verified options and the approved fallback workflow. Voice assignment is a production decision, not a guarantee that a particular number or type of voices exists in every account.
Required Fields Per Video
- Video ID and batch ID.
- Book title and author.
- Intended audience and objective.
- Verified source facts.
- Scene ID.
- Target duration: 10 seconds.
- Approved voiceover text.
- Assigned narration voice.
- TikTok Symphony visual prompt.
- Negative prompt and prohibited elements.
- Book cover or approved reference assets.
- Visual-to-voiceover relevance check.
- Actual rendered narration duration.
- Actual scene duration.
- Factual accuracy status.
- Branding and artifact inspection status.
- Correction notes.
- Final export path and review status.
- Author-delivery approval.
Per-Video Production Tracker (complete all 15 rows)
| Field | V01 | V02 | V03 | V04 | V05 | V06 | V07 | V08 | V09 | V10 | V11 | V12 | V13 | V14 | V15 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Video ID | |||||||||||||||
| Batch ID | |||||||||||||||
| Book title | |||||||||||||||
| Author | |||||||||||||||
| Intended audience and objective | |||||||||||||||
| Verified source facts | |||||||||||||||
| Scene ID | |||||||||||||||
| Target duration (10 seconds) | 10s | 10s | 10s | 10s | 10s | 10s | 10s | 10s | 10s | 10s | 10s | 10s | 10s | 10s | 10s |
| Approved voiceover text | |||||||||||||||
| Assigned narration voice | |||||||||||||||
| TikTok Symphony visual prompt | |||||||||||||||
| Negative prompt / prohibited elements | |||||||||||||||
| Book cover / approved reference assets | |||||||||||||||
| Visual-to-voiceover relevance check | |||||||||||||||
| Actual rendered narration duration | |||||||||||||||
| Actual scene duration | |||||||||||||||
| Scene count (1 preferred, 2 maximum) | |||||||||||||||
| Factual accuracy status | |||||||||||||||
| Branding and artifact inspection status | |||||||||||||||
| Correction notes | |||||||||||||||
| Initial generation asset path | |||||||||||||||
| Final export path (distinct from initial generation) | |||||||||||||||
| Review status | |||||||||||||||
| Author-delivery approval |
Release Gate (every video must pass before author delivery)
Use explicit statuses only: PENDING, FAILED, CORRECTION_REQUIRED, or PASSED.
| # | Check | Status | Notes |
|---|---|---|---|
| 1 | Correct book and approved source information | ||
| 2 | Correct scene count: one preferred, two maximum | ||
| 3 | Correct 10-second final scene duration | ||
| 4 | Correct voiceover text and 10-second audio duration | ||
| 5 | Visuals demonstrably relevant to the narration | ||
| 6 | Appropriate voice selected and pronunciation reviewed | ||
| 7 | Book cover accurate and not materially distorted | ||
| 8 | No unintended branding, interface debris, or hallucinated text | ||
| 9 | No unsupported plot claims or materially misleading imagery | ||
| 10 | Successful final export, with the entire rendered video watched | ||
| 11 | Approved correction notes and final status | ||
| 12 | Explicit QA approval before delivery |
Rule: A failed or unchecked video must never be represented as author-ready. If a check cannot be completed automatically, perform a human review step and do not mark it passed just because a file exists.
Status Definitions
PENDING�?" not yet reviewed.FAILED�?" reviewed and failed at least one check.CORRECTION_REQUIRED�?" defect identified and correction assigned.PASSED�?" all 12 checks passed and human QA approval recorded.
Deliberately Faulty Example (training)
A 10-second TikTok video for A Siren Sings shows attractive storm footage while the voiceover says, "The heroine sacrifices herself to save the lighthouse keeper." The approved source synopsis does not contain that event. The visual is also generic ocean footage with no link to the book.
Detection and rejection:
- Compare the voiceover to the master brief and approved synopsis.
- Mark the unsupported plot claim as a Gate A critical defect.
- Mark the generic visual as a Gate B/C defect (irrelevant B-roll).
- Status:
FAILED/CORRECTION_REQUIRED. - Correction: rewrite the voiceover to a supported claim or a clearly fictionalized visual that makes no factual claim; regenerate or replace the footage within the correction reserve.
- Re-run all 12 release checks before any author delivery.
Notes
- Keep initial generation assets and final approved exports in separate folders/paths so the two are never confused.
- Where a platform imposes a non-removable watermark or licensing restriction, flag the limitation honestly. Do not claim the output is compliant if it is not.
- The book cover is the only permitted branding within the generated video.