Suno has recently launched a significant update, v5.5, for its AI music model, emphasizing user control and customization. While previous updates primarily focused on improving fidelity and vocal naturalness, v5.5 introduces three key features: Voices, My Taste, and Custom Models.
According to Suno, the Voices feature, the most requested by users, enables them to train the vocal model using their own voices. This feature allows users to upload clear vocals, finished tracks with backing music, or sing directly into their devices’ microphones. To maintain authenticity, users must also speak a verification phrase, although this safeguard may potentially be circumvented by existing AI models imitating celebrity voices.
Once trained, the Voices feature enables users to have an AI version of themselves sing on uploaded music or AI-generated compositions. Custom Models further enhance personalization by allowing users to train Suno on their music. By uploading a minimum of six tracks and assigning a custom model name, users can tailor v5.5 responses to prompts.
Moreover, the My Taste feature learns and adapts to users’ music preferences over time, incorporating favored genres, moods, and artists when generating styles using the magic wand tool.
While My Taste is accessible to all users, Voices and Custom Models are exclusive to Pro and Premier subscribers, offering advanced customization options for music enthusiasts.
Source: The Verge