localize-a-video
Localize video through translated meaning, timing, voice, captions, graphics, metadata, and cultural context. Use when an approved master must serve audiences in another language or locale.
ソース情報
- リポジトリ
- protosphinx/sphinxstack
- ソースの最終更新活動
- 2026年7月29日 13:55
- 検出された SKILL.md の言語
- 英語
- スター
- 3
- フォーク
- 0
インストール方法
デフォルトでは、最初にソースを確認する Prompt が選択されています。直接コマンドに切り替えるか、ローカルコピーをダウンロードすることもできます。
ソースファイルを確認
インストールを決める前に、SKILL.md と SkillsMP に表示されている付属ファイルをお読みください。
ファイルエクスプローラー
2 ファイルSKILL.md を表示中
SKILL.md
ソースの指示 · 読み取り専用プレビュー- name
- localize-a-video
- category
- media
- description
- Localize video through translated meaning, timing, voice, captions, graphics, metadata, and cultural context. Use when an approved master must serve audiences in another language or locale.
# localize-a-video
Preserve intended meaning and audience experience rather than translating isolated words.
## Procedure
1. Define target locale, audience, distribution, dialect, accessibility, rights, and approval owner.
2. Lock and transcribe the source with timecodes, speakers, terminology, and on-screen text.
3. Create a context-rich translation with glossary, names, numbers, humor, idioms, and disputed terms.
4. Choose subtitles, dubbing, voice-over, graphics replacement, or a combination.
5. Adapt line length, timing, lip constraints, reading speed, layout, and safe areas.
6. Record and mix voices with consent, casting, pronunciation, and performance direction.
7. Replace graphics, dates, units, URLs, legal text, credits, and platform metadata.
8. Review linguistically, culturally, technically, and in the final player.
9. Deliver locale-labeled masters, captions, transcript, and a localization report.
## Guardrails
- Never imitate a person's voice without clear authority and disclosure.
- Do not erase culturally meaningful context to force literal timing.
- Keep source and localized claims aligned and fact-checked.
## Done
- A localized delivery package contains approved video, captions, transcript, and metadata
- Meaning, terminology, timing, voices, graphics, rights, and playback are verified
- Native-language review findings and final versions are recorded
GitHubで見る