AI Karaoke Maker — Turn Songs into Sing-Along Videos
This is a browser-based karaoke tool that strips vocals from a song, times lyrics to the audio, and lets you edit both before exporting a karaoke video, WAV backing track, or LRC/SRT lyric file. Automatic detection and separation cost processing credits, while manual editing and local export are free.
- Uploads are capped at 10 minutes and 25 MB, and automatic lyrics cost one credit per input minute while vocal separation costs four, so a five-minute track uses 25 credits for both.
- You can skip automation entirely by importing your own instrumental and an LRC or SRT file, then editing timing, line breaks, and highlight colors manually.
- Video renders locally in the browser as MP4 (WebM fallback) at 720p or 1080p, and the site warns that dense mixes can leave vocal bleed, so you should listen to the separated audio before performing.
This is a browser-based karaoke tool that takes an uploaded song and turns it into a sing-along video. You can upload MP3, WAV, M4A, FLAC, MP4, or WebM files up to 25 MB and ten minutes long. Automatic lyric detection costs one processing credit per started input minute, and vocal separation costs four, so a five-minute track runs 25 credits for each. Editing, previewing, and exporting don't use credits, and you can skip the AI steps entirely by pasting your own lyrics or importing an LRC or SRT timing file and using an instrumental you already have. Video renders locally in the browser, offered at 720p or 1080p, in MP4 with WebM as a fallback, and you can also export a WAV or the timed lyric file.
The workflow has a few distinct stages. You can edit detected lyrics, correct names and accents, split or merge lines, and nudge word timing while listening, which matters because the page itself admits detection is a draft, especially on dense or reverb-heavy mixes. Vocal separation may leave traces, so the advice is to listen to the separated backing track before you rely on it. Choose a highlight color, font size, alignment, and a custom image or video background, then pick landscape, portrait, or square output. A guide vocal slider lets you practice a tricky phrase and fade it out as you get comfortable. Drafts stay in the browser, and a verified account can save one complete project per mode up to 64 MB for 30 days.
The one caveat the page leaves easy to miss is rights. The site says removing vocals or adding timed words does not grant any music rights, so the uploader is responsible for the song, the background image, and wherever the finished video gets posted. For a quick, low-cost way to build a practice or party video from a track you own, the manual route is free, and the AI route is useful as long as you plan to check the timing by ear.