AI Video Lectures Built From Your Sources

A short video is usually ready in about 3 to 12 minutes, and longer lectures take longer: a 45-minute lecture can take an hour or more. It runs on Scholarly's servers, so you can close the tab and come back to the finished video.

70+ output languages and regional variants, including English, Spanish, French, German, Portuguese, Italian, Dutch, Czech, Polish, Greek, Hebrew, Ukrainian, the Nordic languages, Hindi, Chinese, Japanese, Korean, Arabic, Thai, and Turkish. The narrator is told which language it is reading rather than guessing from the text, so each language is spoken natively, numbers and abbreviations included, and the captions and transcript are generated in the same language. Arabic, Hebrew, and other right-to-left lectures are laid out right to left, not just translated. The source and the lecture do not have to share a language.

There is no timeline editor. To change a lecture, adjust the sources, custom instructions, mode, length, or language and generate a new version.

Yes, to start. The Free plan includes one lifetime AI Creation Credit, shared across every Scholarly tool, and it can make one narrated video lecture of about 2 minutes at 720p with Scholarly branding, in any mode, with no credit card. Watching and sharing by link are free; MP4 download, 1080p, a custom length up to 45 minutes, and 15 AI Creation Credits a week come with Premium.

On the Free plan, a video lecture is a fixed length of about 2 minutes. On Premium and Laureate you set a target length up to 45 minutes before generating (15 minutes is the default), and the lecture is planned to that length. The finished lecture shows the length you asked for next to the runtime delivered. Each video costs 1 AI Creation Credit with the default model, plus 1 extra credit above 15 minutes and 2 extra above 30 minutes. The Short video toggle targets 30-45 seconds in portrait format with any style.

Yes, on any paid plan. Premium and Laureate download the finished lecture as an MP4, and turning the Scholarly logo off when you generate gives a watermark-free file; the export pack adds the companion slides PDF (Standard mode), the script, and caption files. On the Free plan you watch the video in Scholarly and share it with a view link that opens in the browser, but download is not included.

A natural voice explains your material over new visuals, with chapters, captions, a searchable transcript, optional question checkpoints and AI chat. Choose Standard, Standard Fast, Illustrated, Math Lecture, Doodles or Freestyle. Standard Fast is the default. Full lessons are landscape; the Short video toggle makes a focused portrait video in any style.

Yes. You don't need a source file. Type a topic like "Krebs cycle", "linear algebra eigenvectors", or "causes of the French Revolution" and Scholarly writes the script, builds the animations, and narrates the full lecture from scratch.

Yes - those are the two most common inputs. Use PDF to Video to convert lecture slides or a textbook chapter into a narrated video lecture, or Notes to Video to turn typed or handwritten study notes into one. Both feed the same generator, so you get the same chaptered, narrated video with a transcript, AI chat, and one-click flashcards.

Scholarly is built for learning from your own material. It reads your PDF, notes, or topic as a teaching source and renders an animated, chaptered lecture up to 45 minutes with a study loop around it: interactive question checkpoints, AI chat, and one-click flashcards on a spaced-repetition schedule or graded quizzes from the same source. Free includes one 2-minute video, so you can judge it on your own material.

An AI video lecture is a narrated lesson that the AI writes and builds from material you give it: a PDF, notes, slides, a link or a topic. In Scholarly it is a video with one narrator voice, visuals that build in step with the narration, chapters, captions, a transcript and question checkpoints. It is not a recording of a person teaching, and there is no avatar.

PDF, Word, PowerPoint, OpenDocument, RTF, Markdown and plain text files, spreadsheets, images, and audio or video files, plus YouTube links, web articles, pasted text or a topic. You can also pick files from Google Drive or Dropbox, and pages from Notion. Audio, video and YouTube sources are read through their transcript.

No. Scholarly lectures are narrated visuals: slides, diagrams, illustrations and figures from your source, with one narrator voice. There is no avatar, talking head, lip-sync or voice cloning, so pick an avatar tool if you need a presenter on screen.

When you give it sources, the script is written from the files and pages you select, and you can add instructions about what to cover. AI can still misread or oversimplify, so check the transcript against your material before you rely on the lecture or share it.

Yes. Captions switch on in the player and follow the narration, in the same language as the lecture. In the Illustrated style the captions are part of the picture. On paid plans the export pack includes .srt and .vtt caption files to use with a downloaded MP4.

Yes, for lectures in the Standard and Standard Fast styles. They come with a companion slide deck in the Slides tab, and paid plans export it as a PDF, a PowerPoint file, or to Google Slides with Google Drive connected.

Yes. Copy the embed code from the share dialog and paste it into the page; the app lists Moodle, Canvas, Notion, Confluence and SharePoint as places it works. The link has to be open to anyone who has it, with no password, and the embed then plays for anyone who can see that page. This is an embed, not an LMS integration, and there is no SCORM or xAPI export.

Yes. A finished lecture can be narrated again in another language for 1 AI Creation Credit on the Auto model, and the versions are linked from the lecture page. This works on lectures made in Scholarly. An uploaded video cannot be dubbed or translated.

Gemini Notebook (formerly NotebookLM) makes Video Overviews of your sources in three formats: Cinematic, Explainer, and Short. Google's help page gives a length only for Short, about 60 seconds. A Scholarly video lecture is about 2 minutes on Free, and on Premium and Laureate you set a target length up to 45 minutes, with chapters, captions, a transcript, question checkpoints, and an MP4 download with .srt and .vtt caption files. Choose Gemini Notebook for a quick overview and Scholarly for a full-length lecture.

Yes. Upload the slides, PDF chapter, or notes you teach from, choose a length, language, and voice, and Scholarly generates a narrated, chaptered lecture video with captions, a transcript, and question checkpoints. Premium and Laureate set any length up to 45 minutes, render in 1080p, and download the MP4 with .vtt and .srt captions for your LMS. Turn on Use my script as written to keep your exact wording. There is no avatar or on-screen presenter.

Yes. Upload a .pptx deck or a PDF export of your slides. Scholarly reads each slide and its speaker notes, writes the narration (or keeps your script word for word when Use my script as written is on), and renders a narrated lecture with chapters and captions. You pick an AI narrator voice in any of 70+ languages and regional variants; it is not a recording of your own voice.