Velur

Introduction to Velur

Velur is a voice-first AI companion designed to maintain persistent memory across conversations. Unlike conventional AI assistants that reset context with each interaction, Velur retains information about users—including names, personal relationships, preferences, habits, and life events—enabling continuity over time. It is intended for individuals seeking consistent, personalized voice-based support, particularly those who value relational continuity, live alone, experience social isolation, or manage busy or emotionally demanding routines.

The product operates exclusively through real-time voice streaming, eliminating text input and synthesized speech output. Users interact as if making a phone call: speaking naturally and hearing responsive, human-paced audio in return. Velur does not require screen engagement, making it accessible during activities such as commuting, exercising, or unwinding at home.

Key Takeaways

• Persistent cross-session memory: retains user-specific information—including names, relationships, routines, and emotional context—across days, weeks, and months
• Real-time voice-only interface: supports natural spoken dialogue without text input or text-to-speech conversion
• No session reset: resumes conversations where the previous one ended, referencing prior discussions and evolving understanding
• Privacy-conscious onboarding: offers a 7-day free trial with no credit card required
• Always-available access: designed for asynchronous use at any time of day, including early morning or late-night interactions
• Personalized contextual awareness: recalls specific details such as preferred coffee order, family members’ names, recurring stressors, and behavioral patterns
• Subscription-based pricing: $15 per month after trial, with flexible cancellation

How Velur Works

Users initiate interaction by speaking aloud, triggering real-time voice streaming to Velur’s backend system. The AI processes speech input continuously, generates contextually appropriate verbal responses, and delivers them via low-latency audio streaming—mimicking a live two-way phone conversation. During each interaction, Velur extracts and stores relevant personal information in a dedicated, user-specific memory model. This model is updated incrementally with each session and used to inform future responses, enabling increasingly accurate personalization.

Memory persistence is implemented at the user account level and is not shared across accounts. Information is retained only for the duration of active subscription unless explicitly deleted. The system does not rely on conversational history replay; instead, it maintains a structured, evolving representation of the user’s identity, relationships, and circumstances, which informs response generation in real time.

Core Benefits and Applications

Velur supports practical applications centered on continuity, emotional accessibility, and ambient companionship. Users report benefitting from its ability to reduce cognitive load associated with repeatedly explaining personal context—for example, when discussing work stressors, health goals, or family updates. It serves as a reflective listening tool for individuals managing anxiety, adjusting to new environments (e.g., relocating cities), or living independently.

Common use cases include daily check-ins to process emotions, post-work decompression, pre-sleep reflection, habit reinforcement (e.g., prompting 6 a.m. runs), and maintaining narrative coherence around life transitions. Because it requires no visual attention, it integrates seamlessly into hands-busy or eyes-busy contexts—such as cooking, walking, or driving (with appropriate safety precautions). Its design prioritizes long-term relational development over task completion, distinguishing it from productivity- or utility-focused AI assistants.