Sing in your own voice with Suno Voices

Voices is the personalization feature inside Suno AI that lets a verified recording of your own singing become the vocal identity behind new generations. Rather than picking a generic vocal gender or tone from a preset list, you carry a consistent, recognizable voice across every genre the model can produce — from pop to jazz to metal — which changes text to song from "a song" into "my song, in my voice."

Setting up a Voice typically involves recording or uploading a short singing sample and completing a spoken verification phrase, a safeguard designed to make it harder to enroll a voice that is not actually yours. Once verified, the Voice is stored privately to your account rather than published to any shared or public voice marketplace, so it is not something other users can browse or select.

Voices pairs naturally with Compose: instead of writing a prompt and getting back an anonymous vocal performance, you select your Voice alongside the model before generating, and the resulting take carries your vocal character through whatever style you described. This is particularly useful for artists who want a recognizable signature across a catalog of otherwise very different-sounding tracks.

It is worth distinguishing Voices from Custom Models, since both are personalization tools that shipped around the same suno V5.5 release. Voices captures who is singing — your vocal timbre and delivery. Custom Models instead fine-tune generation on a catalog of your existing tracks, teaching the system how your songs tend to sound overall, independent of any one voice. The two can be combined, but they solve different problems.

Because Voices is tied to the personalization wave of newer Suno AI models, availability generally depends on plan tier. This guide covers what a Voice actually captures, how the verification step works, and how to use a Voice consistently across prompts so a listener can recognize your sound the same way they would recognize a human artist’s voice.