More screenshots

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIDeepSeekDeepSeekMuseMuse+22 more

    Lyric Caption Video – Word-Synced Karaoke Lyrics

    1

    Turn any song + lyrics into a word-by-word synced karaoke lyric video (MP4) or .ass captions, 100% local and free. YouTube & Shorts formats, Arabic/RTL supported.

    $7

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    Lyric Caption Video – Word-Synced Karaoke Lyrics

    Lyric Caption Video – Word-Synced Karaoke Lyrics

    Example session with this skill installed

    Make a YouTube lyric video from song.mp3 using the lyrics in lyrics.txt (English). Yellow highlight, show the next line underneath, use cover.jpg as the background.

    • Read your context and instructions
    • Compiled the lyric caption video

    Summary

    Aligned 38 lyric lines (214 words) to song.mp3 and rendered a 1920×1080 karaoke lyric video.

    StepOutput
    Alignmenttimings.json, 214 words timed, avg drift < 0.1 s
    Subtitleslyrics.ass, karaoke sweep, #FFD400 highlight, next-line preview
    Videolyric_video.mp4, 3:42, H.264/AAC, YouTube-ready

    Commands run

    python scripts/align_lyrics.py song.mp3 lyrics.txt -o timings.json --language en --demucs
    python scripts/make_ass.py timings.json -o lyrics.ass --next-line --bold
    python scripts/render_video.py song.mp3 lyrics.ass -o preview.mp4 --bg cover.jpg --preview 30
    python scripts/render_video.py song.mp3 lyrics.ass -o lyric_video.mp4 --bg cover.jpg
    

    Next steps

    • Watch preview.mp4. If captions feel late, re-run make_ass.py with --offset -0.15
    • For a Shorts version: --res 1080x1920 --position center in both commands
    • Import lyrics.ass into CapCut/Premiere if you want to edit further
    timings.json

    Connects securely to your tools. The creator never sees your data.

    About this skill

    Give your agent a song and its lyrics, and get back a lyric video where every word lights up exactly when it's sung.

    How it works

    1. Align – forced alignment (stable-ts / Whisper) times your lyric text to the audio. Nothing is re-transcribed, so spelling and line breaks stay exactly as written.
    2. Style – generates a karaoke .ass subtitle: smooth sweep, per-word highlight or plain lines; your colours, font, position, optional next-line preview.
    3. Render – burns it onto an image, looping video or solid colour with ffmpeg → MP4 ready for YouTube.

    Why it's useful

    • Runs entirely on your machine: no subscriptions, no uploads, no watermarks
    • 16:9 for YouTube and 9:16 for Shorts / Reels / TikTok
    • Proper Arabic, Hebrew and Persian support (right word order, UTF-8 BOM)
    • Export just the .ass to keep editing in CapCut, Premiere, DaVinci Resolve or Aegisub
    • Built-in troubleshooting for drift, offsets, fonts and Windows installs

    Built from my own workflow for producing cover and lyric videos.

    Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Video Skills.

    Requires local FFmpeg and Python 3.9+. GPU (CUDA) strongly recommended for speed. Does not fetch lyrics or audio from the web.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before buying a skill.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    Money-back guarantee

    Doesn't do what its page promises? You get your money back.

    Trust & safety

    Security scan

    Security scan passed 9 days ago

    Accounts
    GitHub
    Software
    Python

    Permissions

    Terminal / Shell
    Read Files
    Write Files

    Allowed Hosts

    api.github.com
    Listed9 days ago

    What's inside

    Frequently Asked Questions

    Unlike standard AI transcription which can hallucinate or change words, this skill uses 'forced alignment' to ensure the output matches your specific lyric sheet exactly, word-for-word, with millisecond precision.

    Yes, the skill is fully compatible with Arabic, Hebrew, Persian, and Urdu, correctly handling right-to-left text direction and word ordering in the generated video and subtitle files.

    The skill requires a local environment with Python 3.9+ and FFmpeg installed; it includes a dedicated setup script to handle the installation of stable-ts, Torch, and optional vocal separation tools.

    You will receive the complete suite of alignment and rendering scripts, pre-configured styling templates for various social media formats, and a comprehensive guide on how to integrate the output into professional NLEs like DaVinci Resolve.

    While this skill generates high-quality rendered videos, it also exports industry-standard .ass subtitle files, giving you full control to adjust fonts, colors, or positioning in external editors like Aegisub or CapCut.

    Yes, the skill includes a vocal separation feature powered by Demucs, which isolates the singer's voice to ensure the AI can accurately time words even against loud or complex instrumental backing tracks.

    New in Image & Video