According to CoinMeta, the audio company AI has completed a seed round of financing worth tens of millions of yuan, with Jinsha River Venture Capital as the latest investor. Unlike most audio models, Noiz does not intend to continue focusing on music generation, speech recognition, or TTS. Instead, they aim to enable AI to understand the physical world behind sounds. For example, the model will not only recognize that "the door is closed" but also need to know the material of the door, the force applied, the distance sound travels, and how the structure of the room affects the sound. Noiz is specifically collecting data that includes information about space, material, distance, and direction. The ultimate goal is to build a "sound version of a world model" that allows the model to infer the sounds that should be heard based on the current environment and events, thereby enabling machines to understand what is happening around them through sound.