Language Swap

Translates and dubs video into another language in one call while preserving the original speaker's voice. The dubbing worker handles transcription, translation, voice cloning, and A/V sync server-side, so you're not chaining together separate transcribe/clone/TTS steps or managing duration drift yourself. Lipsync is on by default (fal sync-lipsync, works up to 5 minutes, about $4/min). You can also dub different sections of a video into different languages using a segmented plan, where the same cloned voice speaks every language and uncovered ranges keep the original audio. Captions are burned in by default, and you can request bilingual subtitles with the translated speech on top and original below. Skip lipsync with a flag if there's no on-camera face or you want to avoid the cost.

skills.sh installs: 1.2K
GitHub stars: 41
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add pika-labs/pika-plugins --skill language-swap --agent claude-code

Installs into .claude/skills of the current project.

Files

/pika:language-swap

Tools below are Pika MCP tools, named bare — call each under whatever prefix your session exposes for the Pika MCP.

Translate and dub a video into another language while preserving the original speaker's voice. Pipeline: dub (one worker call) → lipsync (default ON) → burn target-language captions or bilingual captions.

The dubbing worker does the heavy lifting in a single call: it transcribes, translates, preserves each speaker's voice server-side (no separate clone step), and returns a fully A/V-synced video — so there is no manual transcribe/clone/TTS/replace chain to manage and no duration-drift handling to do by hand.

Segmented / multi-language dub (per-range languages)

Use this when the user wants different languages on different parts of one video (e.g. first half Spanish, second half Japanese), or wants to translate only some sections and keep the rest in the original voice. Both are the same thing: a timeline of segments, each tagged with a language; any uncovered range keeps the original audio.

dub_video takes a segments plan instead of target_language (pass exactly one — they are mutually exclusive):

SKILL.md on GitHub