Meituan’s LongCat‑Video‑Avatar 1.5 replaces its audio encoder with Whisper‑Large, cuts inference to eight steps, and, after a 770‑person, 13,240‑rating evaluation, outperforms competing models in lip‑sync, style generalization, multi‑person scenes, and overall visual fidelity.
Li Mu (沐神) returns to Bilibili after a year to showcase Higgs Avatar v1, a fully AI‑generated real‑time digital human that can listen, speak, lip‑sync and display facial expressions, with performance metrics showing 16 ms per frame on a single H100 GPU and potential applications ranging from customer service to training, while also raising ethical considerations about identity and trust.
LongCat Video Avatar 1.5 is now open‑source, delivering commercial‑grade lip sync, physical realism, long‑video stability, multi‑person interaction and 15× faster inference through Whisper‑large audio encoding, DMD 8‑step distillation and LoRA adapters, and it outperforms leading closed‑source models in extensive human‑rated benchmarks.
Pixelle-Video lets users input a single topic and, within minutes, automatically generates a short video with script, AI‑generated visuals, synthesized voice‑over, background music, and final rendering, eliminating the need for manual editing or specialized software.
The article details how JD’s JoyStreamer and JoyStreamer‑Flash models overcome text‑control weakness, multimodal conflict, and identity drift to enable long‑duration, free‑state, real‑time interactive digital‑human video generation, surpassing current SOTA models in benchmark scores and reaching 30 FPS inference speed for e‑commerce live streaming.
Baidu’s award‑winning script‑driven multimodal digital‑human technology, recognized at the 2025 World Internet Conference, showcases breakthroughs in real‑time multimodal coordination, high‑fidelity video generation, and cost‑effective live streaming across e‑commerce, education, and legal sectors.
The MIDAS framework introduced by the Kling Team combines autoregressive video generation with a lightweight diffusion denoising head to deliver real‑time, high‑quality digital‑human synthesis under multimodal control, achieving sub‑500 ms latency, 64× compression, and robust performance across multilingual dialogue, singing, and interactive world modeling tasks.
Kuaishou's Kling-Avatar leverages a multimodal large‑language‑model‑driven two‑stage generation framework to produce minute‑long digital‑human videos that synchronize lip movements, facial expressions, and body gestures with audio, achieving high visual quality, identity consistency, and controllable storytelling across diverse scenarios.
During a pre-conference interview, AI expert Wang Mingzhong details how multimodal AI dressing, video résumé creation, short‑video templates, and interactive digital‑human live streams are technically realized for 58 Home Services, highlighting model training, workflow optimization, and future fusion of template‑based and agent‑driven video generation.
This guide introduces GenHuman—a uni‑app and Webman based digital‑human solution—covers its cross‑platform compatibility, provides a Linux environment check, and walks through a step‑by‑step deployment using BT panel, nginx configuration, database import, and admin access.
This article details the end‑to‑end design of Taobao Live's AI digital human system, covering six core components such as LLM‑driven content creation, interactive dialogue, TTS voice synthesis, visual synchronization, audio‑video engineering, and a scalable backend, while also discussing product evolution, automation challenges, and future roadmap.
This article presents a comprehensive overview of the AI‑driven digital‑human live‑streaming solution used by Taobao, detailing six core components—including LLM‑based content generation and interaction, TTS, visual driving, audio‑video engineering, and backend services—while sharing architectural diagrams, cost‑reduction strategies, productization insights, and future directions.
This article presents a comprehensive practice summary of building an intelligent digital‑human system, covering six core modules—LLM content generation, LLM interaction, TTS synthesis, visual driving, audio‑video engineering, and backend services—while detailing data collection, signal processing, ASR annotation, speaker clustering, model optimization (V1‑V4), evaluation metrics, and future research directions.
This article outlines the end‑to‑end architecture and practical solutions behind creating intelligent digital humans for live commerce, covering LLM‑driven content generation, real‑time lip‑sync, image‑driven avatar creation, automated material review, lightweight model training, and a roadmap toward fully automated, high‑performance virtual presenters.
This article details the end‑to‑end workflow for creating intelligent digital humans for live streaming, covering large‑language‑model‑driven content generation, multi‑stage TTS architecture, extensive audio‑signal processing, speaker clustering, front‑end text normalization, back‑end acoustic modeling, and quantitative evaluation of model improvements.
In this CVPR 2025 showcase, Ant Group presents three cutting‑edge papers—EchoMimicV2 introducing an open‑source semi‑body digital human generation framework, RivuletMLP offering an efficient MLP‑based architecture for compressed video quality enhancement, and a quantized super‑resolution model that achieves real‑time 3× upscaling on mobile NPUs.
This article details the end‑to‑end AI pipeline for creating intelligent digital humans in live streaming, covering LLM‑driven script generation, multimodal data integration, error‑prone number handling, DPO fine‑tuning, experimental results, and future directions for more human‑like presentations.
The article explains what digital humans are, reviews the evolution of 3D modeling techniques—from early 2D hand‑drawn methods and 3DMM to deep‑learning‑based implicit models like NeRF—and discusses current challenges and future research directions.
This article explains how AI-powered digital humans can modernize banking by offering modular, multi‑modal interaction, personalized multilingual service, 24‑hour availability, and risk‑aware automation, while detailing the underlying AI foundation, decision engine, visual rendering, and deployment strategies.
This case study details how 58租房 tackled cumbersome landlord publishing by redesigning the workflow with smart location (LBS), AI‑driven shooting assistance, OCR‑based document recognition, and digital‑human guidance, achieving up to 90% faster operations, higher accuracy, and stronger privacy protection.
EchoMimicV2, an open‑source project from Ant Group's Alipay AI team, introduces an end‑to‑end audio‑driven framework that generates high‑quality semi‑body portrait videos by jointly coordinating audio, pose, and image inputs, while addressing challenges of condition complexity, model stability, and computational cost.
EchoMimicV2 is an open‑source digital‑human framework that generates high‑quality half‑body animation videos from a single reference image, an audio clip, and a hand‑gesture sequence, addressing challenges of facial portrait limits, complex condition injection, and inference latency in audio‑driven animation.
This article describes Shujia Technology's use of artificial intelligence to generate marketing images and videos, outlining the background, challenges of high-volume content production, detailed solutions for image and video assets—including layout models, diffusion models, and digital human synthesis—and future research directions.
EchoMimic, an open‑source project from Ant Group, presents a flexible, audio‑ and pose‑driven digital human generation pipeline that combines 2D, 3D and AIGC techniques, reduces production costs, achieves real‑time inference, and includes a detailed architecture, related work analysis, and future research directions.
EchoMimic is an open‑source AI tool that animates portrait photos into speaking videos using audio or facial landmarks, built on Stable Diffusion with a specialized Denoising U‑Net architecture, and comes with step‑by‑step setup instructions and example demos.
This guide explains the architecture of an interactive digital‑human system, walks through 3D avatar creation with Unreal Engine, details the AI controller that combines ASR, LLM and TTS, and provides step‑by‑step instructions for deploying the open‑source Fay project.
Taobao Life 2’s new AI‑driven “photo‑face” feature automatically converts a single portrait into a stylized 3D avatar in under five seconds by using a 3D morphable model, lightweight MLP mapping, and fine‑grained attribute classification, cutting manual sculpting time from half an hour to seconds while preserving user‑specific details.
The article explains how Ant Group’s Galacean engine enables millions of participants to become digital torchbearers for the Asian Games by leveraging Web3D, physics‑based rendering, low‑code animation tools, and extensive cross‑device testing to create a scalable, interactive 3D experience.
This step‑by‑step guide shows how to capture facial photos, process them with Metashape, create a 3D model, and import it into Unreal Engine's MetaHuman system to generate a realistic digital human without any cost.
This guide walks through building a 3D digital human using Unreal Engine’s MetaHuman Creator, driving it with live facial capture from an iPhone, and applying AI‑based face swapping (roop) to replace the character’s face with Mr. Bean, covering all required tools, setup, and export steps.
This article presents a comprehensive overview of the Music XR Maker framework, detailing how music‑driven AI techniques enable digital human creation, dance generation, lip‑sync, and expressive performance, and discusses data pipelines, model architectures, 3D rendering, product integration, and real‑time deployment within Tencent Music’s Tianqin Lab.
This comprehensive guide walks through the complete pipeline for creating a 3D virtual digital human—covering material shading, UDIM baking, PBR texture setup, facial and body rigging with Faceit and Auto‑Rig Pro, and animation using Mixamo, Atorcore and NVIDIA Audio2Face.
This article presents the Music XR Maker framework and its four core components—music-driven system architecture, dance generation, lip-sync driven by singing voice, and expressive singing facial animation—detailing data sources, AI generation pipelines, 3D rendering, product applications, and future research directions.
At NVIDIA GTC 2023, Kuaishou engineers unveiled cutting‑edge solutions ranging from video quality assessment and enhancement, 3D digital‑human live streaming, a custom TensorRT‑based performance framework, large‑scale recommendation model acceleration, to multimodal massive‑model deployment for short‑video scenarios.
At NVIDIA GTC 2023, Kuaishou engineers presented cutting‑edge solutions ranging from video quality assessment and enhancement to digital‑human live streaming, custom performance‑optimization frameworks, large‑scale recommendation model acceleration, and multimodal massive‑model deployment for short‑video applications.
This article details how the Oasis Engine team tackled the technical and artistic challenges of building a large‑scale, multiplayer, third‑person virtual park for the Alipay Fortune event, covering architecture, scene splitting, asset loading, custom materials, digital human integration, networking, behavior management, and extensive performance optimizations.
This article describes the severe shortage of sign‑language teachers worldwide, presents user demographics, outlines the challenges of bidirectional sign‑language translation, and details the cloud‑native AI architecture, data pipeline, and real‑time recognition and synthesis techniques behind the virtual digital human "Sign Language Translator".
This report presents the background, methodology, evaluation model, results, and key observations of the 2022 digital human system basic capability assessment, highlighting technical, engineering, and security challenges, industry standards development, and future work to advance digital human technologies.
This article provides a comprehensive overview of Baidu's Smart Cloud Digital Human platform, detailing its evolution since 2019, core AI-driven architecture, platform components such as persona management and business orchestration, various industry solutions, and technical Q&A on rendering, latency, and deployment.
The paper presents a cloud‑native architecture for AI‑driven digital‑human live‑streaming, detailing three‑layer asset, interaction, and media modules, real‑time script and Q&A scheduling, fault‑tolerant rendering and control services, and demonstrates how virtual anchors can deliver continuous, lifelike 24/7 e‑commerce streams.
This report presents AliMe MKG, a multimodal knowledge graph designed for live e‑commerce, detailing its business background, construction and application, the three types of multimodal knowledge (triples, sentences, and visual media), the underlying extraction techniques, and its deployment in digital‑human anchors and intelligent live‑room assistants.
This article reviews Alibaba's digital‑human (virtual avatar) research over the past few years, covering the product’s evolution, a six‑stage pipeline for building digital humans, solutions to key challenges in realism, multimodal interaction, and the open‑source MMTK algorithm library.
This article reviews Alibaba's XiaoMi digital human technology, covering its evolution since 2019, a six‑stage pipeline for building avatars, methods to enhance emotional, textual, vocal, and motion expressiveness, and approaches for improving long‑term interactive capabilities such as controllable script generation, multimodal QA, sign‑language translation, and intelligent behavior decision, culminating in the release of the MMTK multimodal algorithm library.
The article surveys digital‑human technology—covering visual form, motion and AI‑driven intelligence—its fast‑growing market, a three‑layer solution stack (hardware/software base, AI platform, application services), end‑to‑end avatar creation workflow, rendering and animation techniques, web‑deployment challenges, and future prospects such as deeper AI, XR/6G and metaverse integration.
This article explores the classification of digital humans, examines how they transform user experience, and presents a detailed design practice for AI‑driven virtual bank tellers, covering visual identity, interaction flow, ergonomic screen layout, and future considerations for natural multimodal interfaces.
The article reports on the seventh China International 'Internet+' College Student Innovation and Entrepreneurship Competition, highlighting its new industry challenge track focused on virtual IP, the award-winning solution from Beijing Institute of Technology using AI-driven digital human production, and Tencent Advertising's virtual human creative service for marketing.
