Empower Your Electronic Devices with AI Voice Chatbot
Empower hardware & software for human-like language conversation
We do research on AI generative models and how to align them with human values.
Utilizing deep learning models such as Convolutional Neural Networks (CNNs), Recurrent Neural Networks (RNNs) and Transformers to enhance the accuracy and efficiency of audio processing tasks. These include tasks like emotion analysis, speech recognition, text-to-speech synthesis and AI music generation.
Research in the field of Large Language Models (LLMs)
Introducing models pretrained on large-scale datasets into image and video tasks to enhance performance in specific domains. These approaches aid in achieving better image and video processing results, especially in situations with limited data.
Research in AI Imageology
Built upon large-scale models, the versatile and controllable text-to-image system intelligently generates images in various styles based on natural language. It offers API services for developers with image creation needs, providing inspiration and high-quality illustrations for content creators.
Is there a limit to the size or complexity of the
Our tool can inpaint areas of any size or shape, as long as they are contained within the image. However, larger or more complex areas may require more processing time and may produce less realistic results.
