# AI Explainer Video Generator — Tutorials with Narration | PV30 > Turn a topic and an outline into a structured explainer with narration: script, per-shot framing, diagrams of record, voice-over and a rendered cut. Canonical: https://www.pv30.com/explainer/ From a topic and an outline to a narrated, structured explainer. An AI explainer video generator takes a topic and an outline and produces a narrated video that follows them in order: script, per-shot framing, voice-over and a rendered cut. PV30 treats the explainer as a structured project, so step 3 of your outline stays step 3 of the film and can be re-cut without touching the rest. Presenters, sets and props are kept as elements with reference images, which is what makes a series viable — the same presenter and the same room appear in every episode instead of being re-described from scratch each time. ## Facts - **Typical length:** 30–300 s · default 90 s - **Inputs:** Topic · Outline · Audience · Duration - **Also start from:** Your own script, or a reference video - **Voice:** Separate voice-over track from source audio - **Languages:** Chinese or English audio & subtitles ## Who it is for Educators turn a topic and an outline into a structured explainer with narration, without booking a studio. Product and enablement teams produce onboarding and how-to videos, then regenerate a single step when the interface changes instead of re-recording the whole thing. ## What you give it The creation form asks for the topic, the outline, the audience and the duration. The assistant writes a well-structured script from that, following your outline rather than inventing its own order. If you already have the script, paste it and PV30 imports it unchanged and parses it into shots and elements. ## How an explainer is built Script → elements and shots → storyboard design → images, video and voice → timeline → final cut. Voice-over is generated as its own track, kept separate from source audio so you can re-time narration without re-rendering picture. Each shot carries beats, framing, camera angle, performance, lighting, lines and duration. Review the design, then revise from the notes before anything is paid for. ## Keeping a series consistent The presenter, the room and any props are elements with reference images reused across shots and across projects, so episode 6 still looks like episode 1. Regenerating never overwrites — you bind the version you want. ## Updating an explainer later Change one line, one shot or one asset and regenerate only that. Text edits never silently regenerate paid media, and anything affected by an upstream change is flagged stale rather than changed behind your back. ## Questions ### How long can an AI explainer video be? The tutorial and explainer type runs 30–300 seconds with a 90-second default, built from multiple storyboard shots on a timeline. ### Can I supply my own outline or script? Yes to both. Give the topic and outline and the assistant writes the script from them; or paste a finished script and PV30 imports it exactly as written and parses it into shots and elements. ### Is the narration a separate track? Yes. Voice-over is generated by MiniMax TTS as its own audio track, separate from source audio, so you can adjust narration independently of the picture. ### Can I reuse the same presenter across episodes? Yes. Presenters, sets and props are elements with their own reference images, reused by every shot that uses them and copyable across projects. ### Which languages are supported? Films can carry Chinese or English audio and subtitles, with an optional story locale. The PV30 app interface is in Chinese and English. --- Screenshots and frames on this page come from demo projects made in PV30; brands and characters are fictional. Results depend on the models and settings used.