编辑“︁Veo”︁（章节）

== 發展 ==
2024年5月，在[[Google I/O]]2024大會上，Google發布了一款名為Veo的{{tsl|en|Multimodality|多模態}}影片生成模型。<ref name="Wiggers 14 May 20243">{{cite web |title=Google Veo, a serious swing at AI-generated video, debuts at Google I/O 2024 |url=https://techcrunch.com/2024/05/14/google-veo-a-serious-swing-at-ai-generated-video-debuts-at-google-io-2024/ |website=TechCrunch |date=14 May 2024 |last1=Wiggers |first1=Kyle}}</ref>根據官方說法，該模型能夠生成超過一分鐘長的1080p 高畫質影片。<ref name="Wiggers 14 May 20243" />同年12月，Google推出Veo 2，透過VideoFX平台提供使用。該版本支援[[4K解析度]]影片生成，並具備更精確的物理理解能力。 <ref>{{Cite news |last= |first= |date=2024-12-17 |title=Google unveils improved AI video generator Veo 2 to rival OpenAI's Sora |url=https://www.thehindu.com/sci-tech/technology/google-unveils-improved-ai-video-generator-veo-2-to-rival-openais-sora/article68994621.ece |access-date=2024-12-20 |work=The Hindu |language=en-IN |issn=0971-751X}}</ref>

2025年4月，Google宣布Veo 2開始在[[Gemini (聊天機器人)|Gemini]]上提供給進階使用者使用。<ref name="未命名-20251117160429">{{Cite web |title=Google launches Veo 3, an AI video generator that incorporates audio |url=https://www.cnbc.com/2025/05/20/google-ai-video-generator-audio-veo-3.html |website=CNBC |date=2025-05-20 |language=en |last= |first= |access-date=2025-05-20}}</ref><ref name="未命名_2-20251117160429">{{Cite web |title=Veo 3 can generate videos — and soundtracks to go along with them |url=https://techcrunch.com/2025/05/20/googles-veo-3-can-generate-videos-and-soundtracks-to-go-along-with-them/ |website=TechCrunch |date=2025-05-20 |language=en-US |last=Wiggers |first=Kyle |access-date=2025-05-20}}</ref>同年5月，Google發布Veo 3，不僅能生成影片，亦可產生與畫面同步的音訊，包括對白、音效與環境音。<ref name="未命名-20251117160429"/><ref name="未命名_2-20251117160429"/>Google 並同時推出名為 Flow 的影片創作工具，該工具結合Veo與[[Imagen]]技術驅動。<ref>{{Cite web |title=Google has a new tool just for making AI videos |url=https://www.theverge.com/news/670181/google-deepmind-ai-videos-app-flow-veo-3-2-imagen-4-io-2025 |website=[[The Verge]] |date=May 20, 2025 |last=Peters |first=Jay |url-status=live |archive-url=https://web.archive.org/web/20250520193230/https://www.theverge.com/news/670181/google-deepmind-ai-videos-app-flow-veo-3-2-imagen-4-io-2025 |archive-date=May 20, 2025 |access-date=May 20, 2025}}</ref>

Veo 3的一項重要創新，在於能生成與影片高度契合的音樂與語音。<ref name="Wiggers 20 May 20252">{{cite web |title=Veo 3 can generate videos — and soundtracks to go along with them |url=https://techcrunch.com/2025/05/20/googles-veo-3-can-generate-videos-and-soundtracks-to-go-along-with-them/ |website=TechCrunch |date=20 May 2025 |last1=Wiggers |first1=Kyle}}</ref>Google DeepMind執行長[[Demis Hassabis|傑米斯·哈薩比斯]]表示，此次發表標誌著 AI 影片生成正式走出「無聲電影時代」。<ref name="Wiggers 20 May 20252" />