Suno dropped v5.5 today, March 26, 2026. The update introduces three features that together make Suno feel more personal than any previous version: Voices, Custom Models, and My Taste.
Each one addresses a different layer of personalization. Voices lets you sing on your own tracks. Custom Models lets Suno learn your sound. My Taste adapts the tool to your creative patterns over time. Here is exactly what each one does.
VOICES
Personas is now renamed Voices. If you had Personas before, nothing is lost. All your previous Personas are waiting in the new Voices tab, ready to use.
The big addition is that you can now record or upload your own audio and use it to sing on your Suno creations. No studio setup required. No complicated mic chain. You give Suno a sample of your voice, and it uses that voice on the tracks you generate.
This was the most requested feature Suno has ever received, and it is easy to see why. Previously, creators who wanted their own voice on a Suno track had to use third-party tools like LALAL.AI or Kits.AI to isolate and transplant vocals. Now it is native to the platform.
A few things to know:
- Each creation using Voices costs 4 credits during the beta period
- Available to Pro and Premier subscribers only
- Your Voice is private by default. However, if you publish a song or share it by direct link with Remix or Cover enabled, others may be able to create from it using your Voice. This is important to understand before you share anything publicly.
CUSTOM MODELS
Custom Models is a different kind of personalization. Instead of applying your voice to a track, you are training Suno on your actual music catalog.
The process is straightforward: upload at least 6 tracks from your own catalog, name your model, and wait a few minutes. When it is ready, you have a personalized version of Suno that knows your sound. It shapes how v5.5 responds to your prompts based on the music you uploaded.
This is not the same as the old Personas feature. Personas (now Voices) applied a specific vocal character. Custom Models goes deeper. It learns your production style, your genre tendencies, your sonic fingerprint, and uses all of that to inform what it generates for you.
Custom Models is available to Pro and Premier subscribers.
MY TASTE
My Taste is the quietest of the three features, but the one that compounds the most over time.
It learns the genres, moods, and references you keep coming back to and uses that to guide your generations. When you hit the Magic Wand in the Styles field, My Taste influences the suggestions you get. The more you create, the more Suno adapts to your musical language.
Unlike Voices and Custom Models, My Taste is available to all users, including the free tier. It works passively in the background, getting better the more you use Suno.
WHAT THIS MEANS FOR CREATORS
Voices is the headline feature. It is what creators have been asking for since the beginning. Custom Models goes further: instead of just reusing a voice, you are now training Suno on your entire sound. My Taste is the quieter feature but it compounds over time. Together, v5.5 moves Suno from a generic generator toward something that actually sounds like you.
If you want to get the most out of v5.5, pairing these new features with strong prompt structure makes a real difference. Our complete guide to Suno lyrics tags covers how to use [Intro], [Chorus], [Bridge] and more to control your output precisely.
For creators who want professional voice cloning that goes beyond what Suno Voices offers natively, ElevenLabs remains the industry leader for high-fidelity voice synthesis across music, podcasts, and audio content.