And the winners are...

gm legends, happy Monday.

Big news first: the 2025 AI Dictation Orbit Awards winners are live, so you can finally see who walked away with the crown.

Here’s today’s lineup: CyberCut cuts down the grind in video editing, Marengo 3.0 takes multimodal video understanding into real-world territory, and FlowLens gives your coding agents a clear view of the bugs you actually want fixed.

Drum roll, please. 🥁

The moment you’ve been waiting for: we’re announcing the winners of the 2025 AI Dictation Orbit Awards. This first Orbit edition is all about the tools that actually showed up in real life, not just on launch day. We looked at reviews, traction, how often people came back, and whether these products quietly became part of daily workflows instead of just living in a “try later” folder.

AI Dictation turned into a seriously competitive space this year, and the final lineup proves it. This Orbit round features Wispr Flow, Willow Voice, MacWhisper, ITO, Alter, Aqua Voice, and Superwhisper. Packed field, very different approaches, same ambition: help people talk more and type less.

Now the fun part. Go see who walks away as the people’s champ, who climbed their way into the rankings over time, and which ones might deserve a spot in your own stack next.

CyberCut helps you turn scripts and long recordings into videos you can actually post. It can build a marketing video from your idea, pull highlights out of long footage, add subtitles without the usual headache, and give you assets when you don’t have your own. The whole thing is built to cut out the chores so you can just make the thing you wanted to make.

🔥 Our Take: Video editing burns time in the dumbest ways, so anything that cuts the busywork is welcome. This isn’t promising genius or artistry, it’s just trying to stop you from wasting half your day trimming footage and fixing subtitles.

Marengo 3.0 is TwelveLabs' biggest update so far. It is a multimodal embedding model that pulls signal from video, audio, text, images and even composed queries like image plus text together. It is built for long clips, fast sports, noisy audio and multilingual content instead of short, polished samples. 

🔥 Our Take: A lot of work in this space chases pretty benchmark numbers on short, clean clips. This goes in the opposite direction and aims at long, chaotic, mixed content that teams actually deal with in production. If your video data looks closer to real life than a demo reel, this is at least pointed at the right problem.

FlowLens lets you capture web app bugs with one click and hand real context to your coding agents. The Chrome extension records what happened in the browser, saves it locally, and the MCP server turns that into a structured replay your agent can work from. Everything runs on your machine, with the option to share flows if you are on a team.

🔥 Our Take: Debugging with agents breaks the moment you have to explain a bug from memory. This gives them something closer to what a developer would want to see, without you copy pasting logs into chat for ten minutes. If you are already using agents for debugging, this plugs a pretty obvious gap.