ngram 2.0 - Create explainers, screencasts, and AI avatar videos
by•
Turn ideas, documents, and recordings into narrated videos. ngram analyzes uploaded videos and screen recordings to identify key moments and build clear product walkthroughs. Create animated explainers or add an AI avatar presenter, then refine your video in chat.
Replies
Hey hunters and builders! 👋
I’m super excited to be hunting Ngram
We’ve all been there: staring at a 40-page research paper, a dense technical document, or a complex product workflow, wishing there was an easier way to digest or explain it. Reading walls of text takes time, and making a proper explainer video manually takes even longer.
That’s why the team built the AI Explainer Video Maker. You just drop in an article, a research paper, a slide deck, or even just a raw topic, and Ngram spins up a fully narrated video with animated scenes.
Whether you're trying to speed up your own learning, break down a tricky product feature for customers, or help your team align on a complex process, Ngram handles the heavy lifting in seconds.
Give it a spin, try plugging in your favorite dense piece of text, and let us know what you think! The team is here all day to answer questions and hear your feedback.
Hey everyone, Anish here, one of the makers of ngram 👋
I get really excited about how much we can learn today. With LLMs, we can explore almost any subject, ask questions freely, and go deeper into things we’ve always been curious about. The opportunity to learn and grow feels tremendous.
What we want to do with ngram is help those ideas click.
You could be trying to understand a research paper yourself, explain a process to your team, or show a customer why the product you’ve built matters. Your starting point might be plain text, a slide deck, or a screen recording.
We want to help you shape that into a story someone can follow and understand.
As a maker, helping someone understand something you care about is a lovely feeling.
Would love to hear what you’re learning or trying to explain. ❤️
Congrats on the launch, Rishi and team! Turning a screen recording into a narrated product walkthrough is especially useful for feature updates. Can ngram keep the video in sync when the product UI changes, or would you create a new one from the updated recording?
@ikoma Thanks, Iman! Today, you’d provide an updated recording when the UI changes.
We’re also launching a GitHub integration soon. The idea is to turn each PR into an explainer video showing what changed in the product, initially for internal teams. That context should also help ngram understand product and UI changes better when creating future videos.
Your suggestion adds a useful next step: making those updates useful for end users too, not just internal teams. We’ll make sure the workflow includes a customer-facing changelog alongside the internal explanation. Really appreciate the feedback!
@ikoma You can also tag @askngram on a text-heavy thread on X/twitter, and it’ll turn the thread into an explainer video.
@ikoma We currently don't keep the video live with updated product changes , but thats coming soon 😀
@ikoma Turing research papers into videos sounds really useful. how do you make sure the narrations and visuals stay accurate to the original materials???
@ikoma @aryan_mehta6 Good question, Aryan. ngram uses the uploaded material to build the script and scenes, and you can review the video and request corrections through chat. For research papers, I’d still check the claims, numbers, and visuals against the original before sharing. Using the paper as a source helps.
You can also tag “askngram” on a text-heavy thread on X/Twitter, and it’ll turn the thread into an explainer video.
Would love to start using this in my demos to show more details about my product, congrats on the launch!
@anas_chhilif Thanks, Anas! A recording of one specific workflow is a good starting point. Tell ngram who’s watching and which details matter, and it can turn that into a narrated walkthrough. Curious which part of your product you’d want to demo first.
You can also tag 'askngram' on a text-heavy thread on X/Twitter, and it’ll turn the thread into an explainer video.
I'd probably use this for all those small product updates that never get a proper demo video. Writing the update is easy, making a video for it is where things get annoying
@nrupen_mandava Exactly, Nrupen. Those smaller updates can be useful to customers, but making a video can feel like more work than writing the update itself.
We’re launching a GitHub integration soon that will create an explainer video for each PR, covering what changed in the product. We’re starting with internal teams and will include customer-facing changelogs too. Your example is precisely where we want this to help.
You can also tag @askngram on a text-heavy thread on X, and it’ll turn the thread into an explainer video.
Seems like a super useful tool. I'm interested in the framing/positioning you guys are going for—the main description centers around product updates (a pain to keep making screen recordings of for every new feature/update), but all the rest of the copy (and the site) frames it as a more general explainer-video-generator.
Regardless, congrats on a cool product and great launch :)
@justlewitt Thanks, Liav! That’s a fair observation. Product updates are a specific problem we’re focused on, but the same workflow also works for explaining a document, teaching a process, or walking someone through a product. This launch brings together animated explainers, screen-recording walkthroughs, and AI presenters. The descriptions could connect those dots more clearly, so I appreciate you calling it out.
Did you have any feedback from people who tried AI avatars? I mean, how do they perceive the concept?
Because e.g. when we used AI-created people, users churn. When we reduced AI, we had more conversions, trials successfully finished etc.
@busmark_w_nika Thanks for sharing that, Nika. I wouldn’t claim that adding an avatar automatically improves conversions, and your results are a good reason to test it with the actual audience. Avatars are optional in ngram; you can make a narrated screencast or animated explainer without one. Was the drop you saw mainly on ads, onboarding, or somewhere else?
ps: You can also tag “askngram” on a text-heavy thread on X/Twitter, and it’ll turn the thread into an explainer video.
@ranjanrishikesh We registered negative sentiment in all places (static images, videos, onboarding process visuals)
@busmark_w_nika Yeah, I think there is still some time left for AI to create indistinguishable human-like video.
@annachianna Thanks, Annalese! Yes, ngram already lets you share a video on a publicly accessible page. You can also update or version that video later without changing the link, so a QR code printed on the packaging can keep pointing to the updated instructions.
Today, you’d have to use a separate QR-code generator with that link. I’ll bring this to our engineering team so we can add QR-code sharing directly to the video page. Really useful suggestion!
You can also tag “askngram” on a text-heavy thread on X/Twitter, and it’ll turn the thread into an explainer video.
Choosing the audience and level of detail before making the explainer makes sense. The same source material needs a different story for someone new to it. Congrats on 2.0, Anish and team!
@oleksii_sekundant Thanks, Oleksii! Someone seeing a feature for the first time needs context that an existing user might want to skip entirely. That’s why we ask who the video is for and how much detail to cover. It should change the explanation itself, not just make the same script shorter.
You can also tag '@askngram' on a text-heavy thread on X, and it’ll turn the thread into an explainer video.
@oleksii_sekundant Thank you! Yes choosing the audience lets us adjust the script and pacing of video accordingly