AI-Powered Features in DaVinci Resolve 20

[ There’s a video demo at the end of this article. ]

The hype surrounding AI presents software developers making tools for creatives with a dilemma. As Dan May, President, Blackmagic Design, Americas, said in our interview at the 2025 NAB Show: “I think the challenge there is, how do you create something that’s going to make life easier for creators without replacing creators? Like, obviously it is not Blackmagic’s goal to replace [the] creators [who] are our customers.”

Recently, Avid coined the term: “Responsible AI,” which is “a proactive and responsible approach to the use of these [AI] tools and technologies for both internal and external applications.” (Avid blog).

Adobe doesn’t use this term, but does follow this approach. Kylee Peña, Adobe’s Senior Product Marketing Manager for the Creative Cloud, described it as creating AI tools that are “commercially safe…, which means they’re trained only on things that we have permission to train on licensed content, public domain, things like that.”

However, “commercially safe” does not mean copyrightable. Lawsuits are still in process seeking to determine whether the derivative work of AI, or the prompts used to create an image, can be copyrighted.

Recently, I reached out to Blackmagic Design, to get a list of the AI features in the current versions of their software that are powered by AI (or, more accurately, machine learning).

Here’s what I learned from their PR team.

DaVinci Resolve is rich with AI powered features. We’ve been including DaVinci Neural Engine or AI-powered features within Resolve since version 16, so this isn’t a comprehensive list of every feature, but is a snapshot at some of the more recent updates.

UltraNR Noise Reduction – a new DaVinci AI Neural Engine driven denoise mode in the Color page’s spatial noise reduction palette. Use it to dramatically reduce digital noise from a frame while maintaining image clarity. Combine with temporal noise reduction for even more effective denoising in images with motion.

AI Audio Panning to Video – the new IntelliTrack AI point tracker lets you automatically generate precision audio panning by tracking people or objects as they move across 2D and 3D spaces. With AI audio panning to video, you can track on screen subjects, accurately generating panning data to match movement against picture!

AI Fairlight FX – the AI based dialogue separator FX lets you rebalance dialogue against background sound and the reverberant sound of the room. You get controls for voice, background and ambience so you can reduce, remix or remove competing sounds. It’s perfect for field recordings and interviews in busy locations.

AI Set Extender – the AI Set Extender creates a scene extension to fill an entire frame based on a simple text prompt. Highlight the area you want to extend, and missing regions caused by limited clip angles, blanking and cropping will be auto generated. You can even create new backgrounds behind foreground objects.

AI IntelliScript – use AI IntelliScript to automatically generate timelines based on the original project script. AI IntelliScript will match the transcribed audio in media clips to the script and construct a timeline of the best selected takes, with any alternative takes placed on additional tracks for editor review.

AI Dialogue Matcher – AI Dialogue Matcher Dialogue Matcher is a powerful tool that automatically matches the tone, level, and room environment of dialogue. You can match audio from two totally different clips for audio consistency. Match dialogue recorded in different environments, on different devices, or on shoots spanning several days.

AI Multicam SmartSwitch – in the edit page, Multicam SmartSwitch automatically switches multi-cam angles based on the active speaker in a scene. After creating a multi-cam clip, click SmartSwitch in the multi-cam viewer. SmartSwitch analyzes and automatically selects clip angles based on audio, and lip movement in the video.

AI VoiceConvert – the revolutionary Voice Convert tool applies a pre-generated voice model to an existing voice recording, retaining its inflections, pitch variation, and emotion. Improve your own voice if recorded in a noisy environment, or create perfect ADR using the original actor’s own voice.

AI SuperScale – now features 3x and 4x enhanced upscaling, ensuring the highest visual quality of all media in a project, regardless of source resolution. Perfect for working with archival material or stock footage that needs to be matched to the deliverable resolution.

AI Magic Mask 2 – updated for even more accurate tracking within a single mode. Use points to select people, objects and regions, and paint tools to refine the selection faster than ever. Magic Mask intuitively tracks the motion of a selected area, even around obstructions and in low quality clips.

AI Resolve FX Depth Map 2 – updated for faster scene analysis and more practical mattes. Quickly isolate foreground characters to make them pop against their environment or apply lens blur to the background of a shot. You can refine the resulting matte and isolate specific depths for grading.

AI IntelliCut – AI IntelliCut Audio Processing provides powerful clip-based audio processing, automating time consuming tasks in seconds. Remove silence removes low level or silent areas for a cleaner audio track. You can split dialogue into a separate track per speaker for individual mixing and create an ADR list to replace dialogue.

AI Audio Assistant – when a timeline has audio elements that are not mixed or balanced, AI Audio Assistant can automatically create a professional audio mix. Audio Assistant organizes tracks, evens out dialogue levels, adjusts sound effects and music to the dialogue track, and automatically creates a mastered final mix.

AI Beat Markers – when editing to the beat, you can use the AI Detect Music Beats function to analyze audio clips containing beat-driven music and automatically place markers to indicate beats. You can use these as visual indicators for cut points, or drag to snap clips to the beat’s markers at the nearest frame.

IntelliTrack AI Tracking – powered by the DaVinci AI Neural Engine, optimizing tracking and stabilization in the color and Fusion pages. It can also be used in Fairlight to track on screen subjects and auto generate audio panning within a scene, saving time and producing more immersive audio. IntelliTrack AI tracking lets you use video to track the Fairlight audio panner and you can mix and master stereo, 5.1, 7.1, and even immersive 3D audio formats including 5th order ambisonics!

Larry adds: Here’s a video demo of how four of these AI-powered tools work: Magic Mask, DeHaze, Relight and Depth Map.

Four Killer Color Effects Only in DaVinci Resolve 20 – Studio [v]

TRT: 4:57 — MPEG-4 HD movie