Kling VIDEO 2.6 introduces simultaneous video and audio generation
Kling VIDEO 2.6 generates video with integrated voiceovers, sound effects, and ambient audio in a single pass, eliminating the need to create silent footage for later dubbing. VIDEO 2.6 generates videos up to 10 seconds long and supports Chinese and English voices.
Kuaishou Technology has launched Kling VIDEO 2.6, a breakthrough model that includes a "Native Audio" feature that enables the generation of video clips with audio in a single workflow. Like its competitor Veo 3, Kling VIDEO 2.6 eliminates the traditional workflow of creating silent footage and adding any audio effects separately.
The "Native Audio" feature supports two modes of generation: traditional text-to-audio-visual generation based on text prompts only, and image-to-audio-visual generation that accepts text prompts alongside reference images as input. Both generation modes enable users to create videos up to 10 seconds long.
Kling VIDEO 2.6 can generate videos with perfectly synced human voices, sound effects, and ambient audio in a single pass. The model supports Chinese and English voice generation and maintains a world-leading position in Chinese voice generation. From a more technical viewpoint, Kling Video 2.6 excels in three critical areas: audio-visual synchronization that tightly aligns voice rhythm with visual motion, high-quality audio output with rich layering that mirrors professional mixing standards, and robust semantic understanding of complex storylines and colloquial expressions.
The technology supports diverse audio types, such as speech, dialogue, narration, singing, rap, and mixed sound effects. This makes Kling VIDEO 2.6 valuable across advertising, social media, and e-commerce applications. Advertisers can generate complete product showcases with narration and sound effects with a single click, while social media creators can produce multi-character dialogues and music performances more efficiently. Kling VIDEO 2.6's release notes showcase several examples of outputs suitable for these and other use cases.
Ellie Ramirez-Camara is the News Editor at Data Phoenix, where she writes the daily AI newsdesk — covering model releases, research, funding rounds, and policy across the AI and machine-learning industry. She tracks announcements from labs and startups alike and distills them into clear, source-linked reporting for practitioners.
More news

AWS releases six open-source Hugging Face deployment skills for SageMaker

Google Research releases MilleMiglia logistics benchmark generator

AWS launches AgentCore Runtime V2 with elastic memory and snapshot starts
