Midjourney has released the AI image generation model V7, adding a “Sketch Mode” that supports voice-to-image generation.

Midjourney has released the Midjourney V7 version of its AI image generation model and initiated alpha testing. The V7 version enhances text comprehension capabilities in regular updates, upgrades the detail texture and quality, improves the coherence of character limbs, hands, and object structures, and enables the “Personalized Model” feature by default, which can learn user preferences. It also introduces a brand-new “Sketch Mode” that supports a conversational interface, real-time editing, and voice recognition generation functions.