Suno v5.5 Adds Voice Cloning and Custom Models
- •Suno v5.5 launches featuring voice cloning, custom model tuning, and personalized listening preferences.
- •New 'Voices' feature allows subscribers to record and verify their own unique singing voices.
- •Professional users can now build custom models by training AI on their original music.
Suno has officially released version 5.5, marking a significant leap in AI-driven music generation by prioritizing human identity and creative control. This update introduces "Voices," a highly anticipated feature that allows Pro and Premier subscribers to capture their own singing voice and integrate it into the generation process. To address security and ethical concerns, the platform includes a verification system where users must record a specific phrase to ensure the uploaded audio matches their live speech, keeping these digital voices private to the creator's account.
Beyond vocal replication, the update focuses on deep personalization through "Custom Models" and "My Taste." Professional users can now customize the v5.5 architecture using their own original tracks, creating a bespoke model that understands and replicates their unique musical style—a process known as fine-tuning. Meanwhile, the "My Taste" feature uses a recommendation-style algorithm to learn a user's favorite genres and moods over time, streamlining the creative workflow for hobbyists and professionals alike.
This release also signals Suno's broader strategy of collaboration with the traditional music industry. By enhancing voice fidelity and sound personalization, the company is laying the groundwork for future models developed in partnership with major labels. These tools aim to position AI as a digital instrument designed to amplify human instinct rather than as a complete replacement for human artistry.