Suno v5.5 Complete Guide: Voices, Custom Models & My Taste Explained (2026)

In March 2026, Suno officially launched v5.5 (official announcement), shipping the three most-requested features from its community: Voices (verified voice cloning), Custom Models (personal fine-tuned models), and My Taste (a personalization engine). This release shifts AI music from a generic singer to something that actually sounds like you.

This guide covers three things: what each feature actually does, how to use them inside SunoMV, and how v5.5 stacks up against Google’s Lyria 3 Pro, released the same week.

What is Suno v5.5? One-sentence positioning

Suno v5.5 isn’t a pure audio-quality bump — it’s an identity upgrade. Songs you generated in earlier versions always sounded like “somebody else’s voice.” Now, per the official Help Center categories:

Voices: verified voice cloning for Pro/Premier subscribers who capture or upload their own singing voice
Custom Models: fine-tune v5.5 on your own catalog — up to 3 personal models per account
My Taste: a preference engine available to everyone, including free users

Together, they give AI music a personal signature for the first time.

Feature One: Voices — Verified Voice Cloning

What it solves

Before v5.5, asking the AI to sing in your voice meant guessing prompt keywords forever. Voices skips that entirely: record a sample, and Suno knows exactly what you sound like.

How it works

According to the v5.5 help article:

Open the Voices panel
Suno gives you a randomly selected phrase to read aloud (live or uploaded)
Suno runs a voiceprint check to confirm the voice belongs to you
Once verified, your Voice is ready to use in future generations

Key constraints:

Pro / Premier subscribers only
Mandatory voiceprint verification (you can’t clone somebody else’s voice)
Designed to suppress unauthorized voice-cloning risk

Using Voices alongside SunoMV

SunoMV positions itself as “the MV layer on top of Suno” — once you’ve generated a song with your Voices on Suno’s own platform, paste the song URL into SunoMV for an instant music video with synced lyrics:

Paste link → auto-fetch audio and metadata
Frame-accurate word-by-word lyric sync
7 subtitle styles + AI-generated visuals
Export at 720p / 1080p / 2K

In short, Voices makes the song sound like you, and SunoMV makes the music video look like you too.

Feature Two: Custom Models — Up to 3 Fine-Tuned Models per Account

What it solves

Voices handles “whose voice,” Custom Models handle “whose style.” If you have a catalog of original tracks, earlier Suno generations always carried a generic “Suno default sound.” Custom Models let you feed your own work back into the model so it learns your style.

How it works

Per the v5.5 announcement and Help Center documentation:

Open the Custom Models panel (Pro / Premier only)
Upload your original tracks (10–60 recommended, the more stylistically consistent the better)
Suno fine-tunes v5.5 into a personalized model
Pick your custom model in the generation flow

Core limits:

Maximum of 3 Custom Models per account
More uploads + more stylistic consistency = better learning
Uploads must be works you hold rights to (copyright matters)

How it connects to the SunoMV workflow

Custom Models live on the Suno side of the house, but the output is still a Suno URL. SunoMV already supports v5.5 as a model option:

In the SunoMV generation page, the model picker shows Suno V5.5 – Custom Models, Voices & My Taste (Pro tier). If you’ve already generated a song in Suno using your Custom Model, just paste the URL and SunoMV will pick up the MV production pipeline from there.

Feature Three: My Taste — A Personalization Engine for Everyone

What it solves

Voices and Custom Models are Pro-only personalization weapons. My Taste, on the other hand, is Suno’s personalization engine for every user, free tier included. It doesn’t change the songs you generate — it changes what Suno recommends and how defaults lean.

How it works

You don’t need to train My Taste manually. As you use Suno normally, it learns:

Genres you replay often
Moods and tempos you come back to
Style combinations you save and share

Those signals get distilled into a “Taste Profile” that nudges recommendations and defaults toward your actual preferences.

Availability: all users. This is the only v5.5 capability with no paywall.

Why SunoMV users care

Most SunoMV users are content creators (YouTube / TikTok / podcast), so they generate a lot of songs over time. Once My Taste has been fed enough signal, every subsequent song is closer to the sound you actually want — meaning fewer regenerations and less wasted credits.

The 4-Step SunoMV Workflow for Suno v5.5

Putting the three features together, here’s the standard flow for SunoMV users:

Step 1 · Select the v5.5 model
On the SunoMV song page, pick Suno V5.5 (Pro tier) in the model menu. v5.5 is currently Suno’s top-quality tier: more natural vocals, broader stylistic range, tighter prompt adherence.

Step 2 · Upload or clone your Voice (Pro users)
Complete Voice verification on Suno’s native platform. Content creators only need to do this once; every future song can carry your own voice.

Step 3 · Let My Taste learn your preferences
No manual action needed — use Suno normally (listen, like, generate), and after a few days the default suggestions will sit much closer to your taste.

Step 4 · Paste the song URL into SunoMV for the music video
Once you have a song you’re happy with on Suno, copy the URL into SunoMV:

Auto-detects and fetches the audio
Word-level lyric sync on the video timeline
Pick a subtitle style + AI visual style
One-click music video export

From text prompt to finished MV in the time it takes to drink a coffee.

v5.5 vs Lyria 3 Pro: do you have to pick one?

Released the same week, Google Lyria 3 Pro is v5.5’s most important competitor. Their positioning is genuinely different:

Dimension	Suno v5.5	Lyria 3 Pro
Core value prop	Personalization (your voice, your style)	Structural awareness (intro/verse/chorus/bridge)
Max length	~4 min (Suno V4.5+ up to 8 min)	Up to 3 min (official)
Voice cloning	✅ Voices (verified)	❌ Not supported
Custom fine-tuning	✅ Custom Models (max 3)	❌ Not supported
Preference engine	✅ My Taste (all users)	❌
Watermarking	—	✅ SynthID watermark (official)

Bottom line: if you need “a song that sounds like you,” v5.5 is the only realistic answer. If you need “structurally clean, commercially safe, compliance-first” production music, Lyria 3 Pro is the safer pick. SunoMV’s advantage is that you don’t have to choose — both models are available in the same model picker, switch as needed.

FAQ

1. Can every Suno user try Voices?

No. Voices is limited to Pro / Premier subscribers and requires voiceprint verification to activate. Free users stay on the default voices.

2. How many songs should I upload for a Custom Model?

Suno recommends 10–60 stylistically consistent tracks. More uploads + more consistency = a better-trained model. Fewer than 10 tends to underfit.

3. How many Custom Models can each account have?

Three. So choose the tracks and target style carefully before training.

4. Is My Taste paid?

No. My Taste is available to all users (free tier included). It’s the only v5.5 feature with no subscription gate.

5. Can SunoMV make videos from v5.5 songs?

Yes. Paste your Suno song link into SunoMV and the pipeline handles lyric sync, AI visuals, subtitle styles, and export at 720p / 1080p / 2K.

6. Is the max length of v5.5 much different from v5?

Roughly the same — around 4 minutes. If you need longer tracks, switch back to v4.5+ (up to 8 minutes).

Wrap-up

Suno v5.5’s three features aren’t about stacking parameters — they’re about giving every creator a personal key:

Voices makes the song sound like you
Custom Models makes the style feel like you
My Taste makes the recommendations understand you

For SunoMV users (especially content creators), the workflow is clear: craft “your version” on Suno, render “your MV” on SunoMV.

Open SunoMV, pick Suno V5.5, and ship your personal music video.

Brought to you by SunoMV · SunoMV Team