近日，OpenAI发布了其最新的文本生成视频模型Sora，该模型能够根据用户提供的提示词生成逼真的视频内容。这一突破性的技术引起了业界的广泛关注，但同时也暴露出了其存在的一个显著问题——生成1分钟的视频需要超过1小时的渲染时间。
根据IT之家的报道，尽管Sora模型在演示中展示了令人印象深刻的生成能力，但网友们的反馈指出了一个不容忽视的技术瓶颈。Reddit社区上的讨论显示，研究人员主要展示了预先选择的示例，并未向公众开放自定义提示词的功能。此外，最长的演示视频时长仅为17秒，这远远无法满足用户对于视频生成速度和质量的期待。
这一现象引发了人们对于视频生成技术实用性的质疑。在信息爆炸的时代，用户对于内容的获取速度有着极高的要求，而Sora模型目前的表现显然无法满足这一需求。长时间的渲染不仅增加了用户的等待时间，也限制了模型的应用场景，尤其是在需要即时反馈的领域，如新闻报道、在线教育等。
OpenAI作为人工智能领域的先行者，其推出的Sora模型无疑代表了当前技术的前沿水平。然而，技术的进步不应仅仅停留在实验室的展示中，更应关注其如何更好地服务于社会和用户。对于OpenAI而言，解决视频生成耗时问题，提升用户体验，将是其未来发展的重要方向。
Title: OpenAI’s Sora Video Generation Model Faces Long Rendering Times
Keywords: Video Generation, Rendering Issues, Technological Challenges
News content:
Recently, OpenAI unveiled its latest text-to-video model, Sora, which has the capability to generate realistic videos based on user-provided prompts. This groundbreaking technology has garnered significant attention in the industry, but it also highlights a notable issue—the time it takes to render a one-minute video exceeds one hour.
According to reports from IT Home, although the Sora model demonstrated impressive generation capabilities in demonstrations, user feedback has pointed out a significant technical bottleneck. Discussions on the Reddit community indicate that researchers mainly showcased pre-selected examples and did not open custom prompts to the public. Moreover, the longest demonstration video was only 17 seconds long, which falls far short of users’ expectations for speed and quality in video generation.
This phenomenon has raised questions about the practicality of video generation technology. In an era where information is abundant, users have high demands for the speed of content acquisition, and the current performance of the Sora model clearly cannot meet these demands. Long rendering times not only increase user waiting times but also limit the application scenarios of the model, especially in areas that require immediate feedback, such as news reporting and online education.
As a pioneer in the field of artificial intelligence, OpenAI’s launch of the Sora model undoubtedly represents the cutting-edge level of current technology. However, technological progress should not be limited to laboratory demonstrations but should focus on how it better serves society and users. For OpenAI, addressing the issue of long rendering times in video generation and enhancing user experience will be important directions for its future development.
【来源】https://www.ithome.com/0/751/364.htm
Views: 2