Super Mode + Smooth Pass: Better AI Lyric Swaps and Vocal Refinement

We just shipped a new feature called Super Mode. It is an optional checkbox in the ChangeLyric dashboard that tells the AI to spend extra time studying your specific song before it starts swapping lyrics.

The idea is simple: the more the AI understands about your exact track — the vocal tone, the phrasing, the energy — the better the output sounds. Standard mode works from general knowledge. Super Mode builds a custom understanding of your song first.

What It Does

When you toggle Super Mode on, the AI spends several minutes analyzing your uploaded song before generating anything. It studies the vocal characteristics, how the singer transitions between phrases, and how the voice sits in the mix.

Then when it generates your new lyrics, it has that deeper context to work with. The result is vocals that match the original more closely — better tone matching, smoother transitions between changed and unchanged sections, and output that preserves the feel of the original recording.

The Time Tradeoff

The first time you use Super Mode on a song, it adds roughly 5-10 minutes of processing time. That is the AI doing its deep analysis.

But that analysis is cached. Every subsequent edit on the same song — different lyrics, different sections, new versions — runs at normal speed. The AI already knows the song. So the real cost is a one-time wait per song.

If you are doing multiple iterations on a song (which you should be), Super Mode pays for itself after the first edit.

When To Use It

Turn it on when the vocalist has a distinctive voice, when the song has complex vocals, or when standard mode results were close but not quite right. For the last few weeks - I've basically ALWAYS USED IT for client work.

Skip it when you are quickly testing lyric ideas, the vocals are straightforward, or you just need something fast for a casual project. Standard mode is still solid for most situations.

How To Enable It

In the dashboard, you will see a Super Mode checkbox when setting up a lyric swap. Toggle it on before submitting. Everything else about your workflow stays the same.

Super Mode is included with your plan — no extra cost. The only tradeoff is that initial processing time on the first run.

Introducing Smooth Pass (Beta)

Once you have generated a Super Mode version, you will see a new panel on the editor page called Smooth Pass. This is a refinement tool — a way to take your AI vocal that is almost there and polish it into something usable.

If you have used Suno before, you know they have a smoothing/refinement step that does this kind of cleanup really well. The problem with Suno is the heavy content moderation — a lot of legitimate creative work gets blocked. Smooth Pass is our answer to that: an open alternative that lets you refine demos without restrictions.

We will be honest — Suno currently does this better than we do. Smooth Pass is in Beta and has some issues following melodies accurately on certain passages. We are actively working to improve it. But if you need to smooth a demo and Suno is not an option for you, this is the alternative.

How Smooth Pass Works

After your Super Mode version is ready, take it into your DAW and build a scratch track — vocals plus instrumental in a single mixed file. Your scratch track is the blueprint. The AI uses it to learn the melody, timing, and phrasing you want.

The closer your scratch track is to your vision, the better Smooth Pass performs. There is no shortcut here — the AI is not going to read your mind. It needs to hear what you want.

Getting Good Results

Get your scratch track as close as possible. Spend time on it. Sing the parts you want changed, splice in sections from earlier versions, do whatever it takes to make the demo sound like the finished product you have in your head.

Run new vocals through a voice changer first. If you are recording your own voice to fill in a section, run it through a voice changer to match the original artist's tone before you bake it into your scratch track. This step makes a huge difference in the final output.

Triple-check your lyrics. The lyrics you submit must match exactly what is being sung in the demo. Lyric mismatches are the number one cause of bad Smooth Pass output. If a word is different, fix it before you hit Smooth.

Optionally pick the vocal gender. There is a toggle for male/female that helps guide the AI when it is uncertain. Use it when the original singer's gender is clear.

What You Get Back

Smooth Pass returns a single isolated vocal track — just the vocals, cleaned up and styled to match the original song. No instrumental, no extras. Drop it into your own mix and you are done.

A few things to keep in mind: the result is not saved anywhere — you have to download it before you leave the page. Do not navigate away while it is processing or you will lose it. And it takes a few minutes per pass.

Why Super Mode Is Required

Smooth Pass only shows up after you have used Super Mode on a song. The reason is simple: Super Mode trains a custom fingerprint of the original artist's voice on that specific song. Smooth Pass uses that fingerprint to anchor the output. Without it, the results are not good enough to ship.