Just eight months after the big public launch of its AI music generator Suno, the company is back with v4 of its powerful song generator.
Suno v4 is designed to be a step up from the original version, with vastly improved sound and some cool new features. This week, I put on my music mogul hat and took a quick look at it to see if the enthusiastic marketing pitch rings true. Match reality.
This new release clearly integrates several things to improve the product. To be honest, Suno had a reputation for substandard sound quality and fairly weak lyrics, so I needed an upgrade.
The good news is that Suno v4.0 is definitely an improvement over previous versions. The sound quality of instruments and vocals is now at least as good as Udio, its main rival for the AI music crown, and the new ReMi (pronounced “ray me”) lyricist is a bit of a treasure in many ways.
Improved lyrics generation
The company is touting ReMi as an edgier lyric creator, and in my testing it’s definitely a refreshing departure from some of the AI-generated, run-of-the-mill crap that both Udio and Suno have produced in the past. It turned out to be a significant change. There are no neons, whispers, or echoes in the mix. I breathe a huge sigh of relief.
Not only is there no lyrical drivel, the real uplift comes from the more natural language used on the track. AI music platforms have historically suffered from bland and cheesy word structures, sometimes making them laughable.
This new lyrics generator produces punchier, less obtrusive lyric phrasing. And it makes a big difference in the final product. However, always be careful with young users, as in some cases it can veer into offensive territory.
In fact, I found it stupidly easy to type in a short five-word prompt, let the lyrics generator do the work, and sit until the platform produced something pretty decent.
Unfortunately, the Suno still has a small issue where it doesn’t know how a song ends, and some of the test tracks I created would suddenly stop mid-phrase for no reason, which was very annoying.
Some issues remain with editing
You should also be a little wary of any lingering awkwardness in song editing. Unlike Udio, which offers rich editing features, Suno still struggles with even basic features like extensions. When you extend a song, you get a new generation instead of adding something to the end of the original track.
This means you have to use an external editor like Audacity to compile your final song, which is a bummer. Editing crop and replace sections is also a bit more complicated than it needs to be, and it’s definitely more complicated than the equivalent feature in Udio.
But on the positive side, it does a really good job of covering songs. In our tests of updating older tracks created in version 3, the new cover generation significantly improved the vocals and instrumentation. I found it to be much better than the remastering feature, which as far as I could tell didn’t seem to offer much improvement in sound quality.
persona is improved
So far, the feature I’ve enjoyed the most is Persona. Creating songs couldn’t be easier: just select a saved genre and mood, and trigger interesting lyrics with simple prompts.
In particular, remember that Suno typically produces full or nearly full three-minute songs from a single prompt.
Also, while the sound isn’t 100% consistent from track to track, it’s still consistent enough to create music that sounds like it came from a single artist. If you close your eyes and squint a little.
addictive distraction
Eventually, I got carried away and started spending hours crafting one-shot prompt songs, and in fact, in just five hours, I had an album of 10 songs from one persona.
I didn’t do any post-generation editing as I wanted to test what the overall quality would be like. As you can imagine, some of the tracks were a bit dodgy to say the least, but considering the lack of effort I put into managing the process, the results were amazing.
The advances in this technology are amazing. verdict? Great upgrade! Persona was a huge hit with this new audio and lyrical quality. In testing, we were able to produce more interesting and likeable tracks than we’ve ever done on either platform.
I can now see a future where these tools will produce a ton of really good, if not great, music. Especially if Udio’s upcoming version 2.0 matches or exceeds this quality.
Are you ready for a world full of amateur works? Find out.
Excuse me, I have a hot K-pop/acid jazz combo right now that needs some stimulation from my brain. Look, I may be a shitty songsmith, but at least I’m happy. That’s important, isn’t it?