Sesame AI has made waves by open-sourcing its revolutionary voice AI, Maya, powered by the CSM-1B model. This advanced speech generation model, with 1 billion parameters, is now freely accessible under the Apache 2.0 license. Known for creating human-like speech from text and audio, the technology uses residual vector quantization (RVQ) for natural results.

This move lowers entry barriers in the voice AI field, empowering developers and smaller businesses to innovate in areas like customer service and accessibility tools. However, while the model unlocks potential for advancements in human-machine communication, it also raises ethical concerns due to its minimal restrictions. Sesame AI emphasizes responsible use, urging users to avoid misuse like voice impersonation or creating misleading content.