近日，全球领先的科技公司Meta公布了一项名为audio2photoreal的AI框架。该框架运用先进的人工智能技术，仅需输入配音文件，就能自动为人物模型进行“对口型”和“摆动作”，从而生成一系列逼真的NPC（非玩家角色）人物模型。
据了解，audio2photoreal AI框架的推出，标志着Meta在人工智能领域取得了重要突破。在此之前，虽然已有相关技术能够生成NPC人物模型，但都需要进行繁琐的后期调整，以使人物模型与配音相匹配。而Meta的audio2photoreal AI框架则能实现自动化的“对口型”和“摆动作”，大大提高了工作效率，节省了人力资源。
meta表示，audio2photoreal AI框架的应用前景广阔，不仅可用于游戏开发，还能应用于电影、电视剧、动画等领域。未来，随着AI技术的不断发展和应用，人们可以期待更多高质量的虚拟人物形象出现在各种作品中。
英文标题：Meta Unveils audio2photoreal AI Framework: Voiceover Generates Photorealistic NPCs
英文关键词：Meta, AI Framework, audio2photoreal, NPC Generation
News content:
Recently, Meta, a leading global technology company, has announced an AI framework called audio2photoreal. This framework utilizes advanced artificial intelligence technology to generate a series of photorealistic NPC (Non-Player Character) models simply by inputting voiceover files, automatically synchronizing the lip movements and actions of the character models.
It is understood that the release of the audio2photoreal AI framework signifies an important breakthrough for Meta in the field of artificial intelligence. Until now, although related technologies already exist for generating NPC character models, they usually require laborious post-production adjustments to match the voiceovers. However, Meta’s audio2photoreal AI framework can automate the process of lip synchronization and action coordination, greatly improving work efficiency and saving human resources.
Meta states that the application prospects of the audio2photoreal AI framework are vast, and it can be used not only in game development but also in film, television, animation, and other fields. In the future, with the continuous development and application of AI technology, people can look forward to more high-quality virtual character images appearing in various works.
【来源】https://www.ithome.com/0/744/255.htm
Views: 1