Suno has announced a major update with the release of version 5.5 of its AI music model. While previous updates improved fidelity and created more natural vocals, v5.5 focuses on giving users more control. The update features three new options: Voices, My Taste, and Custom Models.
According to the release notes, Voices is the most requested feature, allowing users to train the vocal model using their own voice. Users can upload clean acapellas, finished tracks with backing music, or sing directly into their phone or laptop’s mic. Higher quality recordings minimize needed data, and to prevent voice theft, Suno requires a verification phrase, though it could potentially be tricked using AI models of celebrity voices.
After training the Voices feature, users can have an AI version of themselves sing on uploaded music or AI-generated outputs.
Custom Models personalize outputs further by allowing users to train Suno with their own music. Users must upload at least six tracks from their catalog and name the custom model, enabling v5.5 to respond to prompts based on it.
Lastly, My Taste learns user preferences over time, recognizing preferred genres, moods, and artists, and applies them to autogenerated styles.
My Taste is available to all users, while Voices and Custom Models are exclusive to Pro and Premier subscribers.