Recently, Meta AI open-sourced a foundational multimodal language model named SPIRIT LM, which can freely mix text and speech, opening new possibilities for multimodal tasks involving audio and text. SPIRIT LM is based on a pre-trained text language model with 7 billion parameters, which has been continuously trained on text and speech units, expanding into the speech modality. It can understand and generate text like a large text model, while also being capable of understanding and generating speech, and even mixing text and speech to create various forms of expression.

In Pasadena, California, the world's first fully automated AI restaurant - CaliExpress - is set to open in the coming weeks. Located at 561 East Green Street near Madison Avenue, the restaurant is dedicated to driving innovation in the food service industry. CaliExpress collaborates with leading tech companies Miso Robotics and PopID, utilizing advanced automation and artificial intelligence technologies. The kitchen of CaliExpress is equipped with Miso R

Recently, a research team released a framework called HelloMeme, which can hyper-realistically transfer the expression of one person in an image to another person in a different image. As shown in the image below, by providing an expression image (first row), the expression details can be transferred to the character in other images. The core of HelloMeme lies in its unique network structure. This framework can extract features from each frame of driving videos and input these features into the HMControlModule. Through this processing, the research

AI video platform D-ID today announced the launch of two new virtual avatars — Express and Premium+, aimed at enhancing the quality and efficiency of content creation. More and more enterprises are looking to use more human-like AI avatars to ease the burden of video production, particularly in marketing, sales, and customer support. D-ID's new models are designed to secure a place in this competitive market. The training time for the Express avatar is approximately one minute of video.

Adobe announced a series of new features for Adobe Express this week, aimed at providing mainstream users who need to create web pages and graphics with greater integration and creative capabilities. The new features include tighter integration with Creative Cloud applications, as well as new tools for text and shapes. Adobe Express now allows for easier collaboration and content sharing with InDesign and Lightroom. The platform also offers AI functionalities.

Recently, companies like Meta and Spotify have united to publicly criticize the European Union's decisions in the fields of data privacy and artificial intelligence (AI). In an open letter, they pointed out that the EU's decision-making process has become 'fragmented and inconsistent,' which not only affects business operations but could also undermine Europe’s competitiveness in the AI era. Image note: The image is generated by AI, and the image is licensed from Midjourney. The open letter states that with increased regulation on data privacy, businesses face challenges in using data to train AI models.

On September 8, SF Technology launched its vertical large language model for the logistics industry, 'Fengyu', at the Shenzhen International Artificial Intelligence Exhibition. This model showcased practical applications in over twenty scenarios across SF's marketing, customer service, collection and delivery, and international customs sectors, marking a significant advancement for the logistics industry in the field of artificial intelligence.

ChatGPT is set to launch a significant upgrade to its voice functionality, introducing up to 8 unique voices, each with its own distinct code name. These voices can express non-verbal sounds and specific emotions more naturally, enhancing the diversity and immersion of the reading experience. The new voices may be rolled out in batches for greater adaptability, but they have not yet been officially released. These upgrades represent an important advancement in AI voice technology, expected to provide users with a personalized and immersive interaction experience, further solidifying ChatGPT's leading position in the fields of text generation and voice interaction.

Alibaba Cloud's Tongyi Large Model platform has launched a free 'PPT Creation' feature, simplifying the PPT production process through independent research and development. It supports intelligent content generation, beautiful design, and smart arrangement. Users only need to input text or upload documents for the system to automatically generate outlines and create content. The design relies on the Tongyi Wanxiang visual large model to ensure high relevance between images and themes. This feature offers various templates to meet needs such as work reports, academic presentations, and daily sharing. In addition, the Tongyi web version has upgraded its 'Real-time Recording' and 'Reading Assistant' features.

Claude, the AI large model developed by Anthropic, has introduced a LaTeX formula rendering feature that significantly enhances the clarity and professionalism of mathematical expressions. Key highlights include improved consistency in formatting, standardized display of mathematical equations and expressions, making them easier to understand. The incorporation of LaTeX greatly enhances the clarity of answers, especially suitable for scenarios requiring solutions to mathematical problems. This feature meets the high demands of academic and professional fields, allowing users to quickly generate complex mathematical expressions with simple command inputs.

The Gradio tool for LivePortrait has been updated with fine control over animation areas, customizable features for expressions and head poses, and relative motion options that protect privacy. It supports neutral lip expressions, automatically saves, and displays suitable output formats to enhance user convenience. Its core advantage lies in expression transfer technology, which can accurately replicate expressions and dynamics onto different characters or pets, adapting to various creative needs, including facial control for static images and videos, providing users with unprecedented creative freedom.

ByteDance and a research team from Shanghai University of Science and Technology have developed an innovative technology called 'HeadGAP' that can quickly generate highly realistic and animatable 3D virtual head models using just three photos taken from different angles, enabling facial expression synchronization. During the research process, the team first extracted 3D head prior information from a large multi-view dynamic dataset to customize the virtual avatar for the target individual. By utilizing a Gaussian point cloud-based self-decoding network and partial dynamic modeling, in conjunction with inversion and fine-tuning strategies, they efficiently achieved high-quality and diverse...

Peach-9B-8k-Roleplay is an AI model specifically designed for roleplay tasks, fine-tuned from the powerful 01-ai/Yi-1.5-9B model, with a wealth of experience accumulated from over 100,000 dialogue syntheses.

Napkin is a 'visual AI' platform developed by Pramod Sharma and Jerome Scholler, designed to streamline the creative expression process by converting text into visual graphics, helping marketers, content creators, and engineers effectively communicate ideas. The platform generates various visual elements such as flowcharts, charts, and infographics from text input, and allows users to customize icons, colors, fonts, etc. The created visual content can be exported as PNG, PDF, SVG files, or URL links.

In the field of 3D digital humans, research teams from Nanjing University, Fudan University, and Huawei Noah's Ark Lab have proposed innovative solutions to address the issues of multi-view consistency and insufficient emotional expressiveness. They developed the EmoTalk3D dataset, which includes calibrated multi-view videos, emotion annotations, and frame-by-frame 3D geometric information. By constructing a mapping framework from 'speech to geometry to appearance', the research team introduced a new method that enables the synthesis of 3D talking avatars with controllable emotions, significantly enhancing lip synchronization and rendering quality.

Adobe has launched Adobe Express for Education, an AI-assisted platform designed for the educational sector, with new features aimed at providing safe and responsible AI tools for students and educators to enhance the learning experience. The platform includes a real-time assignment system, multimedia presentations, animation capabilities, and themed templates to encourage creative projects. It is free for K-12 teachers and students and compatible with various educational software. New AI features include template generation, image creation, and personalized coloring page tools, without AI text generation to protect students' privacy.

Adobe has recently introduced a new round of generative AI capabilities to its design software, Illustrator and Photoshop, aimed at enhancing the efficiency of the creative process. These updates primarily include Illustrator's Generative Shape Fill tool and Photoshop's image generation feature.In Illustrator, the newly launched Generative Shape Fill tool leverages Adobe's latest Firefly Vector AI model to add detailed vector images to selected shapes through descriptive text prompts.Additionall