The smart AI assistant Doubao, owned by ByteDance, has recently upgraded its text-to-image capabilities. The new feature allows users to generate images containing specified text with a one-click operation.
A relevant person in charge of Doubao's large model team pointed out that Doubao's text-to-image model optimizes its learning ability for Chinese data by integrating LLM (Large Language Model) and DIT (Digital Image Technology) frameworks. This has strengthened its ability to generate Chinese characters, significantly improving the generation results.
It is important to note that at this stage, the feature is limited to adding text in text-to-image generation and does not support generating text by uploading images.
This upgrade demonstrates Doubao's in-depth exploration and continuous innovation in AI technology applications, aiming to provide users with a richer and more convenient intelligent service experience. With ongoing technological advancements, Doubao is expected to play a greater role in the field of smart AI assistants, meeting users' needs in areas such as image generation.
