Kling 2.6 — AI Video with Simultaneous Audio Generation
The first AI video model with simultaneous audio-visual generation. Kling 2.6 by Kuaishou produces synchronized speech, sound effects, and ambient audio alongside 1080p visuals in a single unified pass — no manual dubbing required.
Kling 2.6 — AI Video with Simultaneous Audio Generation
Kling 2.6: Simultaneous Audio-Visual Generation in One Pass
Released on December 3, 2025, Kling 2.6 is the first model from Kuaishou to achieve simultaneous audio-visual generation. Unlike traditional workflows that produce silent video first and require manual dubbing afterward, Kling 2.6 generates synchronized audio and video together in a single pass. Built on a diffusion-based Transformer architecture with a proprietary 3D variational autoencoder, it delivers deep semantic alignment between real-world sounds and dynamic visuals at 1080p resolution and 48 frames per second.
Text to Video with Integrated Audio
Transform text prompts into fully realized videos complete with synchronized audio. Kling 2.6 interprets complex instructions with 15% higher compliance than previous versions, translating detailed scene descriptions into matching visuals and soundscapes.
Image to Video Animation
Bring static images to life with fluid, natural motion. Upload a reference image and Kling 2.6 animates it with precise hand movements, expressive facial details, and full-body motion fidelity suited for dance routines, martial arts sequences, and natural human gestures.
Rich Audio Synthesis
Generate speech, dialogue, narration, singing, rap, ambient sound effects, and mixed audio — standalone or combined. Bilingual speech output in English and Chinese, with other languages automatically translated to English for voice generation.
1080p at 48 Frames Per Second
Output videos at full HD 1080p resolution running at 48fps for exceptionally smooth motion. Choose from 16:9 landscape, 9:16 portrait, or 1:1 square aspect ratios to fit any platform requirement.
Generate Video with Integrated Audio
1. Select Your Input Mode
Choose text-to-video to generate entirely from a written prompt, or switch to image-to-video to animate a reference photo. You can also upload a motion reference clip (3-30 seconds) to guide the movement style of your output.
2. Configure Output Parameters
Set your aspect ratio (16:9, 9:16, or 1:1), pick a duration of 5 or 10 seconds, and decide whether to enable native audio generation. When audio is enabled, specify the type — dialogue, narration, sound effects, singing, or a combination.
3. Generate and Download
Start generation and receive your completed video with fully synchronized audio and visuals. The diffusion-based Transformer architecture processes both modalities simultaneously, so what you download is a complete, ready-to-publish clip.
One Pass: Audio and Video Together
Unified Audio-Visual Pipeline
The defining breakthrough of Kling 2.6: audio and video are generated together in one pass rather than layered sequentially. This produces deep semantic alignment — footsteps match walking, doors slam when they close, and dialogue matches lip movements naturally.
Precision Human Motion
Kling 2.6 delivers blur-free hand movements, accurate finger articulation, and nuanced facial expressions. Full-body movement fidelity captures complex choreography including dance routines, martial arts forms, and athletic sequences.
Accurate Lip Synchronization
Speaking characters exhibit precise mouth-to-audio alignment. Whether generating dialogue, narration, singing, or rap, lip movements track the generated speech with frame-level accuracy across both English and Chinese output.
Motion Reference Control
Upload a motion reference video between 3 and 30 seconds long to guide the movement patterns in your output. This enables uninterrupted motion sequences that follow specific choreography, camera paths, or action styles.
Cross-Shot Character Consistency
Maintain consistent character appearance and identity across different shots and scenes. Characters retain their facial features, clothing, and proportions throughout the generated video.
Bilingual Speech Output
Native speech generation in English and Chinese with natural intonation and pacing. Prompts written in other languages are automatically translated to English for voice synthesis, broadening accessibility for international creators.
Kling 2.6 Video Examples
Explore videos generated by Kling models — synchronized audio, precise human movement, and detailed visual storytelling across diverse scenarios.
Frequently Asked Questions
Create Videos with Synchronized Audio Today
Experience the first AI video model with simultaneous audio-visual generation. Produce 1080p videos with speech, sound effects, and music in a single pass.
Explore Other AI Models
Seedance 2.0
2K cinematic video with native audio
Seedance 1.5 Pro
1080p video generation with audio
Kling 3.0
4K video generation model
Kling 2.5
Turbo Pro video generation
Nano Banana
AI image generation model
Nano Banana 2
Next-gen AI image generation
Nano Banana Pro
4K AI image generation
Nano Banana Edit
AI image editing model
Z-Image Turbo
Ultra-fast AI image generation