embedded-captions
Add captions to an existing talking-head video without editing footage, from plain verbatim to cinematic and VFX styles
66.1k GitHub starsupdated Oct 13, 2026
Loading…
Skill dựng, sửa, làm phụ đề, dịch và lồng tiếng phim, video và hoạt hình bằng AI.
Add captions to an existing talking-head video without editing footage, from plain verbatim to cinematic and VFX styles
Build motion and scene transitions in HyperFrames with GSAP, Lottie, Three.js, Anime.js, CSS keyframes and other adapters
Produce 3Blue1Brown-style explainer videos with Manim: math animations, algorithm visuals and architecture diagrams
Generate narration and voiceovers from text with the OpenAI Audio API via a bundled CLI, including batch runs
Transcribe audio and video files to text and label speakers in interviews or meetings
Transcribe speech, polish and translate subtitles, then burn custom-styled captions into your video
Cut silence, still and black frames from video or audio with the auto-editor CLI and render the result
Transcribe media or live mic to text or subtitles with auto-editor's whisper, then cut video by spoken content
Route you to the right Remotion skill for each task when making videos with React code
给 AI 短剧出分镜：三层结构——段（一次视频生成，≤15 秒）→ 分镜（段内 2–5 秒的剪切，认领剧本节拍） → 分镜图（每切一张关键帧：主分镜图钉 0.00 秒，子分镜图钉各自切点）。 每段自带一条 MiniMax H3 视频提示词（官方口径默认英文、逐镜换行，promptLang 可切中文）：对齐指令和 [Shot k] 切点时刻由分镜结构推导、逐字对账，台词逐字进 <d> 块（写法规范已内化为 references/h3-prompt.md，不依赖外部 skill）。 产出 storyboard.json + Markdown + 单页评审报告（分镜节奏带 / 分集分镜表 / 生成批次单 / 配音对齐单，含导出 JSON）。不出图——交付的是每格的画面提示词和它该挂哪些参考图。 18 道质量门全部由脚本确定性检查（第 18 道 shot-recipe 可选：挂上 shot-recipes 卡库才查，不挂就明说跳过）； export 一键导出 H3 投产包（每段提示词 + 按 Picture 序的分镜图清单）。零依赖、零 API key，用当前会话额度。 Use when asked to 分镜、出分镜、镜头表、切镜、storyboard for AI short drama。
Script and automate DaVinci Resolve and build Fusion compositions, macros and keyframe animations
Color grade in DaVinci Resolve via MCP: correct, match shots, build looks and apply LUTs, CDLs or DRX live or offline
Cut, trim, restructure timelines and generate editorial changelists in DaVinci Resolve via MCP, live or offline
Assemble a short-form rough cut for Reels or TikTok from a folder of vlog or behind-the-scenes footage in DaVinci Resolve
把中文短剧剧本和视觉设定写成有戏剧职责、连续性边界与冻结关键帧提示词的分镜 Markdown。用户提出“拆分镜/设计镜头/做镜头表”“场次视觉计划/调度故事板”“比较导演方案”“写首帧/关键帧提示词”或检查轴线、站位、视线、持物连续性时使用；不生成媒体。
编写或修订可拍摄的中文短剧、漫剧单集 Markdown 剧本，也负责保留作者原文并规范化现成剧本。用户提出“写/改一集短剧”“把大纲写成剧本”“优化场景/对白”“去模板感”“去 AI 味润色”“续写下一集”或提供剧本要求进入后续制作时使用；不负责资产、分镜、媒体提示词或终审。
AI music generation with ACE-Step 1.5 — background music, vocal tracks, covers, stem extraction, audio repainting, and continuation for video production. Use when generating music, soundtracks, jingles, or working with audio stems. Triggers include background music, soundtrack, jingle, music generation, stem extraction, cover, style transfer, repaint, continuation, or musical composition tasks.
Process video and audio with FFmpeg: convert formats, resize, compress, extract audio and prep assets for Remotion
AI video generation with LTX-2.3 22B — text-to-video, image-to-video clips for video production. Use when generating video clips, animating images, creating b-roll, animated backgrounds, or motion content. Triggers include video generation, animate image, b-roll, motion, video clip, text-to-video, image-to-video.
Composite clips and overlay text, captions and lower-thirds on AI videos like LTX-2 or SadTalker using moviepy 2.x
Plan shot by shot, generate storyboards, camera coverage and blocking for multi-shot videos
Design and analyze screenplay scenes and sequences using McKee and Syd Field: beats, subtext, turning points, dialogue
Orchestrate screenplay, stage-play or series projects through each stage from premise to submission, with a story bible
Apply Manim Community Edition best practices when writing Scenes, MathTex, animations and running the manim CLI
Break a finished video into a shot-by-shot table with duration, framing and camera motion, measured via ffmpeg
Upscale and restore video in ComfyUI with per-frame ESRGAN or temporal super-resolution via SeedVR2 and FlashVSR
Build WAN 2.2 text-to-video workflows with hi-lo models, lightning LoRAs, VACE modules and two-pass KSamplerAdvanced
Develop characters, worlds and stories for Higgsfield projects, with character sheets and story bibles before prompting
Write, improve and split prompts for AI video generators like Veo, Kling, Runway and Sora, with storyboards and shot lists
Translate and dub videos into another language with HeyGen, cloning the presenter's voice and re-syncing lips
Generate HeyGen presenter videos via the v3 Video Agent, handling aspect ratio, prompts, avatar and voice selection
Dub audio and video into other languages with the ElevenLabs Dubbing API while keeping the original speakers' voices
Compose music with the ElevenLabs Music API: instrumentals, songs with lyrics, background tracks and jingles
Generate sound effects from text descriptions with ElevenLabs, with looping and duration control
Transcribe audio and video to text or subtitles with ElevenLabs Scribe v2, including meeting recordings
Convert text to speech using ElevenLabs voice AI. Use when generating audio from text, creating voiceovers, building voice apps, or synthesizing speech in 90+ languages.
Convert the voice in a recording to another voice while keeping emotion and timing with ElevenLabs Voice Changer
Translate SRT and ASS subtitle files or embedded video subtitles into any language, keeping timestamps and formatting
Create game-ready Blender animation via MCP: walk, run and idle cycles, combat, camera, Graph Editor and NLA
Enhance video with Topaz Labs: upscale, denoise, sharpen, stabilize, colorize, convert to HDR and create slow motion
Create Manim math animations with optional TTS voiceover and synced subtitles for equations, proofs and algorithms
Burn SRT subtitles into video or mux them as a toggleable track, and mix a dub track over the original audio
Generate a time-aligned TTS voice dub for a video from an SRT, using Volcano TTS for Chinese or edge-tts for any language
Combine multiple recordings of one event into a single MP4, auto-switching cameras by audio with optional picture-in-picture
Write effective prompts for AI video models on Replicate or when building video generation features
Create and edit videos with Google Veo 2 and Veo 3: text-to-video, image-to-video, inpainting and video extension
Automatically duck background music under voice and lift it in the gaps with Sonilo, no manual volume automation
Dub videos into other languages with Sonilo, optionally using your own script and getting a re-timed SRT back