Suno v5.5 adds voice cloning and custom model training for Pro subscribers
Suno has released v5.5 of its AI music generation model, prioritizing user customization over quality improvements. The update includes Voices (voice cloning via user-uploaded recordings), Custom Models (style training on user music catalogs), and My Taste (preference learning), with voice and custom features limited to Pro and Premier subscribers.
Suno has released v5.5 of its AI music generation model, marking a strategic shift toward customization features rather than incremental quality improvements.
Three New Customization Features
The update introduces three capabilities:
Voices allows users to train Suno's vocal model on their own voice by uploading clean acapellas, finished tracks, or recording directly via microphone or laptop. Suno requires users to speak a verification phrase to prevent unauthorized voice cloning, though the company acknowledges this safeguard could potentially be circumvented using existing celebrity voice models. Once trained, users can apply their cloned voice to uploaded music or AI-generated outputs.
Custom Models enables users to train the model on their own music library. Users upload at least six tracks, name the custom model, and then use it to guide v5.5 responses to generation prompts, effectively personalizing the model's style output.
My Taste operates as an ongoing preference learner, tracking user selections across genres, moods, and artists over time. The system applies these patterns when users activate the magic wand autogenerate feature.
Availability and Tier Restrictions
My Taste is available to all Suno users. Voices and Custom Models are restricted to Pro and Premier subscription tiers, making customization a premium feature.
The release notes describe Voices as Suno's "most requested feature," indicating sustained user demand for voice-specific control in AI music generation.
What This Means
Suno is betting that personalization—not marginal quality gains—drives engagement and retention. By tying voice cloning and model training to paid subscriptions, the company creates a differentiated value proposition for power users while maintaining a free tier for basic generation. The verification phrase requirement for voice cloning is a light security measure that Suno acknowledges may not be sufficient against determined deepfake attempts, suggesting the company accepts some risk rather than implementing stricter controls that would complicate the user experience.
Related Articles
NVIDIA expands Cosmos with world model training tools, simulation checkpoints and Data Factory blueprints
NVIDIA has expanded the Cosmos platform with new training tools, simulation checkpoints and Cosmos Data Factory blueprints. The company says Cosmos ranks in the top five on six different open model leaderboards.
Ollama v0.40.9 adds Phonon speech generation on Apple Silicon and d3 decision models
Ollama v0.40.9 can generate speech with Phonon on Apple Silicon, with audio streaming through an OpenAI-compatible /v1/audio/speech endpoint. The release also adds the d3 family of decision models in 2B, 4B, 9B, and 27B sizes and brings Flint to Linux and Windows.
OpenAI opens Decisions API to all customers, cutting latency to 100-180 ms on a reconfigured ChatGPT Luna
OpenAI made its Decisions API available to all customers on Thursday, running a reconfigured ChatGPT Luna that returns choices with confidence ratings in 100 to 180 milliseconds, according to OpenAI. The service competes with TypeSafe's Jev, which ranks higher on two public decision-model indices.
Qianjue launches Polibrain, a world model that predicts task-relevant state changes instead of video frames
Robot-brain developer Qianjue launched Polibrain, a world model that predicts task-relevant state changes instead of future video frames. It is adapted to quadruped, humanoid and wheeled robots, according to Pandaily.
Comments
Loading...