Sesame builds personal agents for curious people. Follow a thought. Work out an idea. Discover something new. Preview available now on iOS. Coming to intelligent eyewear in 2027.
Conversational voice companion — natural-sounding speech with breathing, pauses, and filler.
Two characters: Maya & Miles — distinct personalities tuned for warmth and casual chat.
Real-time streaming — sub-second latency for back-and-forth conversation.
Open weights (CSM-1B) — base voice model released for research and commercial use.
Companion glasses roadmap — wearable device designed around the voice experience.
Building voice-first prototypes for accessibility and ambient computing.
Demonstrating expressive TTS in research and academic contexts.
Fine-tuning the open CSM model for branded character voices.
Exploring conversational interfaces for support and coaching apps.
Product teams researching voice UX for assistants and companions.
Developers who want an open-weight TTS baseline they can fine-tune.
Anyone testing what natural voice AI feels like beyond robotic TTS.