我長期處理中英夾雜的技術型影音內容（教學影片、會議與簡報錄音），很清楚這類素材最頭痛的就是專有名詞和英文縮寫被聽錯。這個 Skill 把我實際在用的整套做法工具化：以 faster-whisper 在本機 CPU 就能轉錄（免 GPU、免雲端 API、免費），並用「轉錄前餵專有名詞提示＋轉錄後批次修正同音錯字」兩道工序把準確率拉高；YouTube 連結則走「先抓現成字幕、失敗自動改抓音訊轉錄」的穩定路線，最後一併產出 SRT 字幕、逐字稿與帶時間戳的重點摘要。全程本機運算、檔案不上雲，適合在意內容隱私又要快速消化長影音的人。 I work daily with mixed Chinese-English technical audio, where jargon and acronyms get mis-heard most. This Skill packages my workflow: faster-whisper runs locally on CPU (no GPU, no cloud API)
Drop in your recording details, get back every terminal command you need — from one-time tool setup to a finished 9:16 short-form video. No coding knowledge required. Copy, paste, done. 把錄音丟給 AI，它幫你生成每一條指令——從工具安裝到最終輸出一支 9:16 短影音，不需要懂程式，複製貼上就能跑。
貼上現有腳本或逐字稿，診斷前三秒、節奏、資訊順序與 CTA，取得 5 個替代 Hook、逐句修改，以及 3 套包含口播、鏡頭、字幕與剪輯提示的完整重拍腳本。
傳送你的 Podcast、訪談、直播或演講影片,就能拿到做好的直式短片:挑出能獨立成立的片段,裁成 9:16,加上逐字字幕,畫面跟著說話的人,音量調平,每支附標題、開場金句和貼文文案。不需要字幕檔。可以直接上傳影片,或貼上檔案直連。
Animated SVGs (CSS keyframes, SMIL, Web Animations API) can't be read by standard video tools — ffmpeg only sees a static image. This Skill solves that by driving a headless browser to scrub the animation frame-by-frame at exact timestamps, screenshotting each one, then stitching the sequence into a real H.264 MP4 with ffmpeg. Handles Figma Motion exports, custom CSS-animated SVGs, looping/repeat duration, resolution/fps tradeoffs, and the MP4-has-no-alpha-channel caveat.
將加密貨幣技術分析自動轉成修仙風YouTube影片。支援K線、SNR、SMC、CHOCH、FVG、OB與POC教學，以及BingX主流幣行情分析；產出英文配音腳本、繁體中文字幕、分鏡、縮圖、影片說明，並在工具支援時生成完整MP4。沒有合格交易機會時會明確建議等待，不報明牌、不保證獲利。
Feeds it raw footage (talking-head clips or silent B-roll) and renders a finished short-form video: punch-in zoom that changes rhythm on every cut, animated captions with numbers and keywords auto-highlighted, a rotating pool of transitions, synthesized whoosh sound effects, and music that auto-ducks under speech. Works with or without subtitles - silent B-roll still cuts on a fixed rhythm instead of sitting static. One Python script, one dependency (ffmpeg), runs fully on your own machine.
