Turn a research paper into a podcast: a real example from a PDF

A review paper on the early Solar System, sent as a PDF link, became a two-host podcast. Hear it, copy the request, and see its cost and our check.

How this was made. Audio and video overviews, slide decks and infographics are made with Google's Gemini Notebook (formerly NotebookLM), on accounts that AutoContent operates. AutoContent has no agreement with Google. Google can limit or stop this access at any time.

The podcast24 min 5 s

AI-generated with AutoContent APIDownload MP3 Full transcript (TXT)

What came out, and what we would change

We sent the paper as a PDF link and asked for a summary for a curious listener: the main question, the evidence, what is still uncertain, and plain words for every technical term. With duration at default, the episode runs 24 min 5 s, twice the 12 minutes that the API docs give as the top of that setting. It follows the order of the paper: molecular clouds, young stars, the disk and its snow line, the growth from dust to planetesimals, meteorites and their dating, the delivery of water, and the young Sun.

The numbers match the paper: molecular clouds below 10 kelvin, more than 300 molecules found in space, a water snow line at 100 to 150 kelvin and a soot line near 500 kelvin, the oldest solids dated to 4,567.30 million years ago, and a young Sun at about 70% of its present brightness. The hosts speak of millions of mega-flares; the paper gives about ten million in the first five million years of a star like the Sun.

One thing we asked for is missing: the hosts promise a part on what remains a mystery, but it never comes, and they answer whether asteroids brought the oceans with "yes, they did", although the paper calls this the favored hypothesis. Three small things go beyond the paper. The hosts say "exactly 4,567.30", but the paper gives an uncertainty of 0.16 million years. They turn 10 kelvin into about minus 440 degrees Fahrenheit, which is correct but not in the paper. And the closing point about the James Webb Space Telescope is their own: the paper names the telescope only for disk spectra. As asked, no author is named.

Send a PDF link with type set to file; the pdf type does not accept links. The episode took 11 min 56 s to generate and used 10 credits. For a shorter brief, set duration to short, or style to brief. The same paper also became a German explainer video in this library.

Transcript excerpt

Speech-to-text of the output, not edited and without speaker labels. It is part of the sample output.

Think about standing next to a massive raging bonfire on a freezing winter night. If you are standing a foot away, it's so hot you're sweating. Right. But as you walk backward away from the fire, eventually you cross an exact invisible line where the heat from the fire completely fades, the freezing air takes over, and the moisture from your breath suddenly turns to ice. Building on that analogy, in the protoplanetary disk, the water-snow line is the exact distance from the young star where temperatures drop low enough, around 100 to 150 Kelvin, for water vapor in the vacuum of space to freeze solid into ice on those dust grains.
But wait, if water freezes at that invisible line, what happens to those delicate carbon molecules if they get pulled too close to the fire? Do they just burn up entirely? They do. Much closer to the star, where temperatures hit around 500 Kelvin, there is another boundary, often called the soot line or tar line. The tar line, okay. Inside that line, carbon-based organic materials just vaporize completely. These invisible temperature boundaries dictate everything about planetary architecture because they completely change what building materials are available. Because you can't build with ice if it's vaporized.

The exact request

The body that was sent to POST /content/Create, unchanged, and the same request as curl.

request.json

{
  "resources": [
    {
      "type": "file",
      "content": "https://arxiv.org/pdf/2607.27543v1"
    }
  ],
  "outputType": "audio",
  "text": "Summarize this paper for a curious listener who is not an astronomer: the main question, the evidence, what is still uncertain, and why it matters. Explain each technical term the first time it comes up. Do not name the authors.",
  "duration": "default",
  "language": "English"
}

curl

curl -X POST "https://api.autocontentapi.com/content/Create" \
  -H "Authorization: Bearer $AUTOCONTENT_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"resources":[{"type":"file","content":"https://arxiv.org/pdf/2607.27543v1"}],"outputType":"audio","text":"Summarize this paper for a curious listener who is not an astronomer: the main question, the evidence, what is still uncertain, and why it matters. Explain each technical term the first time it comes up. Do not name the authors.","duration":"default","language":"English"}'
# then poll until "status" is 100:
curl -H "Authorization: Bearer $AUTOCONTENT_API_KEY" "https://api.autocontentapi.com/content/Status/<request_id>"

How we checked it

Transcript: We transcribed the output with Whisper large-v3-turbo and compared the transcript with the source.
Language: Detected en, requested en.
Loudness: -24.6 LUFS integrated, true peak -5.3 dBTP.
Notes: The physical numbers match the paper. Additions: "exactly" for a dated age with an uncertainty, a Fahrenheit conversion, and the closing remark about the James Webb Space Telescope.
Checked on: 2026-11-02. Failed generations are run again and never published.

More examples

All examples