Search
OpenMOSS Makes MOVA Synchronous Audio-Video Model Open-Source, Attaining Cinematic-Level Audio-Visual Synchronization
1 day ago /
Read about 0 minute
Author：小编
On January 30, 2026, the OpenMOSS team hailing from Shanghai Chuangzhi College, in a joint effort with Mousi Intelligence, unveiled China's inaugural high-performance open-source audio-video model, namely MOVA. This innovative model is capable of delivering end-to-end synchronized audio-visual output, thereby shattering the dominance of closed-source technologies like Sora2 and Veo3. The model is adept at generating 8-second audiovisual clips boasting a maximum resolution of 720p. It successfully meets industrial benchmarks in terms of multilingual lip-syncing and the precise alignment of environmental sound effects. Moreover, being entirely open-source, it seamlessly integrates with domestic hardware platforms, such as Ascend.
-
C114 Communication Network -
Communication Home
7 X 24 Track global technological trends
