Corporate podcast production by API: a real company episode
Corporate podcast production with an API: one company blog post became a two-host episode for marketing teams. Hear it, copy the request, see its cost.
How this was made. Audio and video overviews, slide decks, infographics and quizzes 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 episode
Programmatic SEO After AI Overviews: Source-Backed Content Without Scaled Content Abuse (AutoContent blog, 2026-07-01)
PodcastEnglish3 min 50 s
Transcript excerpt
Speech-to-text of the output, not edited and without speaker labels. It is part of the sample output.
So there are three rules for a safe page. First, it needs a unique source set, like a custom data set or a benchmark. Right, so it can't just be rehashed info from somewhere else. Exactly. Second, it has to solve a unique user problem. And third, it needs a visible reason to exist, maybe an interactive calculator or a checklist. Honestly, if a page cannot name its unique evidence, you just do not publish it.
- Length
- 3 min 50 s
- Credits used
- 10 credits
- Cost on the Starter plan
- €0.29
- Generation time
- 4 min 43 s
- Format
- Podcast
- Job
- Company podcasts
- Language
- English
- Generated on
- 2026-11-02
This output used 10 credits, which is €0.29 on the Starter plan (€29 per month for 1,000 credits). Failed generations use none. Generation time runs from the request to the finished file; it depends on the sources, the options and the queue.
What came out, and what we would change
We sent one post from our own blog, a July 2026 article on programmatic SEO after AI Overviews, and asked for an episode of our company podcast for marketing and SEO teams: the main argument, where scaled content abuse starts, the checklist, and how to plan pages around query fan-out, in a professional tone and without code. With duration at short, the episode runs 3 min 50 s.
It follows the article closely. The useful page test (three follow-up questions), the three things a safe page needs, the point that the policy on scaled content abuse applies to any format, grouping fan-out queries by decision stage, and the advice to measure conversions and URL cannibalization rather than page count all come from the article. The hosts add their own pictures, such as a city swap from Austin to Dallas and page count as a "vanity metric".
One sentence goes further than the article. The hosts close by saying that this work makes AI search point back to you as the primary source; the article says the opposite: no special markup secures a place in AI Overviews, and indexing is never certain. They leave out the article's own workflow section, which holds the code we asked them not to read out.
For a team that compares production agencies, these are the figures: the episode took 4 min 43 s to generate and used 10 credits. A company show of this kind is one request per post; add podcastId to collect every episode in one RSS feed, and listen before you publish, because a host can promise more than your source does.
The exact request
The body that was sent to POST /content/Create, unchanged, and the same request as curl.
{
"resources": [
{
"type": "website",
"content": "https://autocontentapi.com/blog/programmatic-seo-ai-overviews-source-backed-content"
}
],
"outputType": "audio",
"text": "An episode of our company podcast for marketing and SEO teams: the main argument of this article, where scaled content abuse starts, the checklist, and how to plan pages around query fan-out. Keep a professional tone. Do not read out code.",
"duration": "short",
"language": "English"
}curl -X POST "https://api.autocontentapi.com/content/Create" \
-H "Authorization: Bearer $AUTOCONTENT_API_KEY" \
-H "Content-Type: application/json" \
-d '{"resources":[{"type":"website","content":"https://autocontentapi.com/blog/programmatic-seo-ai-overviews-source-backed-content"}],"outputType":"audio","text":"An episode of our company podcast for marketing and SEO teams: the main argument of this article, where scaled content abuse starts, the checklist, and how to plan pages around query fan-out. Keep a professional tone. Do not read out code.","duration":"short","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 -2.7 dBTP.
- Notes
- The argument, the useful page test, the three requirements and the checklist match the post. One overreach at the close (AI search points back to you). Own pictures added; the workflow section is left out.
- Checked on
- 2026-11-03. Failed generations are run again and never published.
More examples
All examples- Podcast · EnglishInternal company podcast: a weekly update made from a changelog5 min 36 s · 10 credits (€0.29 on Starter)
- Podcast · EnglishA culture podcast made from a handbook page on remote work5 min 32 s · 10 credits (€0.29 on Starter)
- Podcast · EnglishAn onboarding podcast for new hires, made from a handbook page4 min 56 s · 10 credits (€0.29 on Starter)