Last month I set myself a challenge: create lyric videos for all 12 tracks on a local artist's debut album before their release date on Monday. I had Saturday and Sunday.
I've been running a YouTube channel focused on independent hip-hop and R&B for about three years. My channel, "Deep Cuts Only," has around 42,000 subscribers. I don't make videos as my main job — I work in logistics — but I take it seriously. The artist, a rapper named Devonte from my city, had been a subscriber and supporter of my channel for two years. When he asked if I could help make lyric videos for his album, I said yes before I thought about what that actually meant.
Twelve songs. Two days. No video editing experience beyond iMovie.
This is the honest account of how it went.
Saturday Morning: Reality Check
My original plan was to use a free version of some timeline-based video editor I'd seen in a YouTube tutorial. I spent two hours Saturday morning watching that tutorial, downloading the software, and then realizing it would take me approximately forever to manually sync lyrics for 12 tracks.
I switched to LyricMV around 10am.
My initial concern was accuracy. Devonte raps fast. Some of his verses are dense — 20+ words in four seconds. I'd used AI transcription tools before and gotten burned by them on rap vocals, especially when the artist uses slang, regional dialect, or ad-libs between lines.
I uploaded the first track and watched the transcription come back.
It wasn't perfect, but it was better than I expected. Maybe 88% accurate on the first pass. The AI handled most of the lyrics well; where it struggled was on ad-libs ("yeah," "uh," "let's go") and a few slang terms specific to our city. I made a note of the corrections needed and moved on to see how the editor worked.
The Timing Editor: Where I Spent Most of My Time
For rap music specifically, lyric timing matters a lot more than it does for slower songs. If the text appears half a second late on a fast verse, it looks completely out of sync. Getting it right took longer than I expected.
My workflow ended up being:
- Upload track, wait for transcription (~2 minutes per song)
- Fix transcription errors while listening through once (~5-10 minutes per track, more for dense verses)
- Adjust timing on lines that felt off (~10-15 minutes per track)
- Pick template, preview, render
The template choice took me maybe two minutes per track. I used three different templates across the album to give each song its own feel — darker templates for the introspective tracks, more energetic ones for the hype songs.
Average time per track: about 25-30 minutes.
Twelve tracks: roughly 5-6 hours across Saturday and Sunday.
That's with breaks, meals, and one hour where I got frustrated with a particularly dense verse and walked away to clear my head.
The Mistakes I Made
Mistake 1: I didn't establish a template system upfront.
I chose templates song by song, which meant I spent time re-deciding for every track. By track 7, I'd created inconsistency across the album. Three tracks looked visually similar, two looked completely different, and the rest were somewhere in between. For a cohesive album release, I should have picked 2-3 templates and assigned them by song type before I started.
Mistake 2: I didn't fix the transcription in one pass.
My workflow was to fix errors as I found them during timing adjustment. But I kept finding new errors late in the process and having to go back. Better approach: fix all transcription errors first, then do a second pass for timing.
Mistake 3: I underestimated ad-lib tracks.
Two songs had a lot of ad-libs — repeated phrases, overlapping vocals, crowd noise in one section. The AI transcribed all of it, which technically made sense, but the result was text appearing too frequently and cluttering the screen. I had to manually delete the ad-lib lines from the lyric editor and keep only the main vocal lines. That added 20 minutes to those tracks.
What Worked Really Well
The batch workflow itself. Even with my mistakes, finishing 12 lyric videos in a weekend would have been impossible with a traditional video editor. The AI transcription, even at 88% accuracy, gave me a starting point instead of a blank slate. That's the difference between a 2-hour job and a 20-hour job.
The preview before rendering. I almost rendered one track with a timing error I'd missed. The free preview caught it. I fixed it before rendering. This saved me from having to re-render and re-upload.
Template variety. Devonte was happy that different songs had different visual energy. He said it felt like the videos matched the music, not just generic templates slapped onto every track.
The Results
We uploaded all 12 videos on Sunday evening, scheduled to go live Monday at noon for the album release.
By Wednesday, the album's title track had 8,400 views. That's not viral, but for an artist with no existing YouTube presence, it was significant. Three other tracks cleared 2,000 views in the first week.
More importantly, Devonte told me that several of his listeners had commented that they'd read lyrics they'd previously misheard, and it changed how they heard the song. That's the whole point of a lyric video.
He's already asked me to do his next EP.
My Recommendations for Batch Lyric Video Projects
-
Establish your template palette before you start. Pick 2-3 templates and decide which songs get which template based on mood. Don't decide per track.
-
Fix transcription first, timing second. Two separate passes is faster than one combined pass, counterintuitively.
-
For rap: expect to spend extra time on dense verses. Budget 30 minutes per track minimum, 45 if the artist raps fast.
-
Preview everything before rendering. The extra 30 seconds to check the preview is worth it every time.
-
Download immediately after rendering. I had one track where I forgot to download and had to re-render later. Minor inconvenience, but avoidable.
I went into that weekend expecting it to be miserable. It was actually kind of fun. There's something satisfying about watching the lyrics snap into sync on a track you've been listening to on repeat all weekend.
Devonte's album is called "Before the City Woke Up." If you're into thoughtful hip-hop, look it up.
Marcus T. runs "Deep Cuts Only," a YouTube channel dedicated to independent hip-hop and R&B. He's been covering the underground scene for three years.
