What is LuniSpark? (What)

LuniSpark turns a video or podcast transcript in SRT, WebVTT, or plain-text form into highlight quotes and ready-to-edit draft posts for Facebook, LinkedIn, and X. Quotes are rendered word for word from the cues you select, each with its transcript timecode attached.

When should you use LuniSpark? (When)

Use it after publishing a video, podcast episode, webinar, or lecture, when you need social posts that quote what was actually said instead of paraphrasing it from memory or inventing a plausible line.

Where can you use LuniSpark? (Where)

Open /lunispark/ in a modern desktop or mobile browser. The transcript is processed on your own device; nothing is uploaded, and no account is needed to use the beta. Open the tool or explore other Lunibyte guides.

How do you use LuniSpark? (How)

Import the transcript file or paste the text, confirm the content is yours to use, tick the highlight cues (or start from the labelled rule-based suggestions), review the verbatim quotes and their timecodes, then copy the drafts or download the Markdown export with citations. Follow the controls in order and use a harmless practice file or scenario first.

Why use LuniSpark? (Why)

It keeps promotion honest and cheap: quotes can never be fabricated because they only render from your transcript, and the AI budget is zero by design because the planned intelligence, LuniSpirit, runs locally on your device instead of renting a hosted model.

Important limitations

The beta is draft-only and never posts for you. Highlight suggestions are rule-based, not AI, until LuniSpirit is ready; files are capped at 1 MB, a malformed timestamp rejects the whole file, and everything lives in memory only, so reloading the page clears the session.

LuniSpark Guide: From SRT, VTT or TXT to Facebook, LinkedIn and X Drafts, Step by Step

This guide walks the complete LuniSpark workflow with a real episode transcript: getting the file, importing it safely, choosing highlights you can defend, and leaving with drafts that quote your content word for word. The whole process takes a few minutes, happens entirely in your browser, and costs nothing.

Before you start: get a transcript file

LuniSpark reads three formats. SRT and WebVTT are the caption files most platforms already generate: YouTube Studio offers caption downloads from the Subtitles panel, most podcast hosts export a transcript or captions per episode, and editors such as Premiere, DaVinci Resolve, and Descript can export SRT or VTT directly. Plain text works too — paste a transcript and LuniSpark treats each paragraph as one cue; you simply lose timecodes, and quotes are then cited by paragraph instead. If all you have is audio, run it through your editor's built-in transcription first and export the captions; LuniSpark starts where a text transcript exists.

Step 1: Import and confirm your rights

Open LuniSpark and either drag the file onto the drop area, choose it with the file button, or paste the text into the paste box. Then tick the rights confirmation: you own the content or have permission to use the transcript. Parsing does not start until that box is ticked — on purpose. The beta accepts files up to 1 MB, which covers roughly ten hours of ordinary captions, and it counts and displays how many cues it found plus the running time of the last cue, so you know immediately that the right file loaded.

If the import fails, the message tells you why in plain language. The most common cause is a malformed timestamp somewhere in a caption file; LuniSpark rejects the whole file rather than guessing, because a quote pointing at the wrong moment is worse than an error message. Fix the timestamp in your editor, re-export, and import again.

Step 2: Pick highlights — yours or the rules' starting set

The cue list shows every caption line with its timecode. Tick the moments worth quoting: the claim, the story beat, the number, the sentence you would put on a slide. If the episode is long, press Suggest highlights (rules) first. The beta scores cues by length and by repeated key phrases and pre-selects up to five. Treat that set as a draft from a fast assistant, not a verdict — the suggestions are rule-based, labelled as such in the interface, and not AI. Untick freely; the final selection is always yours.

Your selected quotes appear beneath the list exactly as they will be used: the verbatim text in quotation marks, with the transcript timecode under each one. Clicking a timecode scrolls the cue list to that moment and flashes it, which is your two-way check — draft to source, source to draft — before anything leaves the page.

Step 3: Review, copy, or download the drafts

Step three builds three drafts from your selected quotes: Facebook and LinkedIn get the full quote set under a short heading with a closing question for your audience; X gets the strongest first quote, trimmed at a word boundary if it would exceed the character limit. Each box shows a live character count, and the X box warns you if an edit pushes it over 280. Edit directly in the boxes — your wording always wins over the template. Then either copy each post with its button, or download one Markdown file containing all three drafts plus a citations section listing every quote with its timecode.

Two habits make the output better. First, add your own first line: the template heading is a placeholder, and a specific opening ("The mistake that cost us a month, in our guest's words:") outperforms a generic one every time. Second, keep the timecodes when you post: they invite your audience to check the moment themselves, and that verifiability is the entire point of the tool.

Good to know

Everything happens in memory on your device. Nothing is uploaded, no account exists, and reloading the page clears the transcript and warns you first — download the Markdown file if you want a record. When our local AI, LuniSpirit, is ready, it will slot into step two of this same guide; steps one and three will not change.

From the Lunibyte team