Video & audio

Voice and audio

Narration, subtitles and music beds, local or cloud, with the metrics that make a voice sound human.

local-audio 523 lines · 32 KB

Inside this skill

⚠️ Engine choice: think it through first
0. Doubao speech synthesis (first choice for Chinese)
1. Cloud synthesis (best for English)
2. Local synthesis (offline / free)
3. Subtitle transcription (always local, never uploaded)
4. Music generation (short clips only)
Key pitfalls
Quality self-check
5. Dialogue / final-cut mixing flow (verified)
6. Doubao BGM generation (Seed-Audio 1.0, via AI MediaKit)
7. Hard human-feel metrics (**measurable — never say "natural enough" from feeling**)
8. Audio-source ledger and the placeholder-audio ban (delivery red line)
9. Language and content re-verification (**do not only verify "was it read correctly"**)
10. Lessons feedback protocol (**mandatory after every project delivery, 5 minutes**)

This page is generated from the skill document itself. The outline below is that document's own structure, translated headings included.

Download the file below — it is a single SKILL.md.
Put it in the skills folder your agent reads, in a directory named local-audio.
Start a message with the trigger from the description, or name the skill directly.