Make an Audiogram With AI

A podcast moment, a moving waveform, word-perfect captions — the standard podcast promo asset, rendered free on your desktop.

An audiogram turns a podcast segment into a shareable video: cover art or a guest photo, an animated waveform proving there's audio worth hearing, and word-by-word captions carrying the actual quote — the standard promotion format for audio shows on feeds that don't play audio. ViralMint renders audiograms locally and free: Whisper transcribes the segment (free, on your machine), captions time to the word, the waveform animates against your artwork, and the export sizes to 9:16, 1:1 or 16:9. Where the AI genuinely helps beyond assembly: the viral-clip extractor listens to the whole episode and ranks the most quotable 30-90 second segments, so you promote the moment most likely to stop a scroll rather than the one you happened to remember.

How to make a audiogram with AI

  1. Let AI find the quotable moment

    Run the episode through the clip extractor — it ranks segments by hook strength, so the audiogram features the 'wait, what?' moment, not minute 43's polite agreement.

  2. Add the visual identity

    Cover art or guest photo as the background, waveform style and color to brand, episode title as the header. Saved as a preset, every episode's audiogram matches.

  3. Caption and export

    Word-by-word captions carry the quote for the 80% watching muted. Export all three aspects — 1:1 for feeds, 9:16 for Reels/Shorts, 16:9 for the channel.

Example prompts to start from

From ViralMint's free prompt libraries — click any card for the full prompt and recommended settings.

Frequently asked questions

Why do audiograms work when audio doesn't autoplay?

The waveform signals 'this is a conversation' while the captions deliver its content silently — the viewer reads the hook, and the strong ones click through to the episode. It's a captions format wearing an audio costume.

Is this really free?

Transcription (local Whisper), waveform render, caption burn and export are all local — zero credits. The optional paid step is the AI clip scoring (~$0.02/candidate) that picks the best moment for you.

Video podcast — audiogram or video clip?

If you have camera footage, a captioned video clip usually outperforms an audiogram. Audiograms are the audio-only show's weapon — see the podcast-to-shorts workflow for the video-podcast path.

Make your first audiogram today

ViralMint is a free, open-source desktop app. No subscription — top up credits and pay cents per generation.

Enlarged screenshot