Web page
Your web page
Paste your web page. Nothing to upload.
←web page
Make a podcast
Turn a web page into a podcast
Reading an article aloud is not the same as making it work in audio.
Free to try, no card. Your web page opens in Ancher.
Readable content, declutteredTitle, author, published dateOutbound links
Kept in parts
Every one of these stays separately addressable instead of collapsing into one block of text.
ScriptSegmentsVoiceLength
Podcast
A narrated version you can listen to instead of read.
Text-to-speech on a web page produces the article with the nested clauses, the parenthetical asides and the sentences that only parse if you can see them. Ancher writes a narrated version instead: one idea per sentence, the figures that were in charts spoken aloud because a listener has nothing to look at, and the source credited at the top rather than in a hyperlink.
The honest version
A listener cannot see the chart or click the link
Web writing assumes a screen. It says as the chart shows and puts the number in the image; it says according to this report and puts the source in a link. Both vanish in audio. Because charts and outbound links come through with the content, the narration can speak the figure and name the source, which is what makes it followable without the page.
One screen of one page, at fig. 3, read two ways.Illustrative. The figure is invented; the gap it falls into is not.
As the chart above makes clear, the difference between the two groups is
Chart in the article — not readThe specific never reaches the draft
Podcast
Chart in the article
14 points, narrowing since 2023
from fig. 3
Lands in the draft, with its timecode
- Figures that appear only in charts are read aloud, since a listener cannot see them.
- Linked sources are named in speech instead of left as an unspoken hyperlink.
- Sentences are rewritten for the ear — one idea each, subject repeated rather than pronouned.
What comes out
Podcast, section by section
A narrated version you can listen to instead of read.
Script
The narration, in sentences a person can say in one breath.
Segments
One idea each, with a target length.
Voice
Pacing notes, including which figures to slow down for.
Length
Total runtime and what to cut first.
Source credit
Publication, author and date, spoken at the top.
Web-page-to-podcast prompt
The link on its own gets a summary. This is the instruction that produces the 5 sections above, in that order, with the rules that keep them honest. Paste it with your web page — in Ancher, or in whatever assistant you already use.
Write a narrated script from this article. Credit the publication, author and date in the opening seconds. Rewrite for the ear: one idea per sentence, repeat the subject instead of using pronouns, no stacked clauses. Speak aloud any figure that appears only inside a chart, and carry its caveat with it. Name cited sources in speech rather than leaving them as links. Break it into segments with target lengths.
The detail, if you want it
How each part of a web page becomes a section
| From the web page | Into | Why |
|---|---|---|
| Inline images | Script | Any figure that exists only in a chart has to be spoken, because as the chart shows means nothing to a listener. |
| Outbound links | Source credit | A hyperlink is silent, so a cited source has to be named aloud or the narration attributes nothing. |
| Title, author, published date | Source credit | Publication, author and date belong in the opening seconds, since a listener has no byline in front of them. |
| Readable content, decluttered | Script | Once page furniture is stripped, what remains can be rewritten for the ear rather than read off the screen. |
| Many pages in one collection | Segments | Several articles in one collection make a briefing episode rather than a reading, with a segment per source. |
Everything Ancher keeps from a web page
Readable content, decluttered
Navigation, ads and cookie banners are dropped, so the collection holds the article rather than the page.
Title, author, published date
A date is what stops a three-year-old post from being read as current.
Outbound links
What a writer links to repeatedly reveals the sources they actually trust.
Inline images
Charts and diagrams in an article are usually the evidence the prose is describing.
Many pages in one collection
A single page is an anecdote; the value of this route appears once several are compared.
The four steps
- Add the web page. Paste any URL. Ancher fetches the readable content as clean markdown and keeps the title, author, publication date and outbound links — this is the route for articles, newsletters, forum threads and documentation.
- Nothing gets flattened. Readable content, decluttered, title, author, published date, outbound links, inline images, many pages in one collection stay separately addressable instead of collapsing into one block of text.
- Ask for a podcast. Ancher fills script, segments, voice, length, source credit using the mapping above, not a generic summary pass.
- Check it before you use it. Every specific keeps a link back to where it came from in the web page, so a claim can be verified without reopening the source.
Where this goes wrong
- Narrating someone else's article in full is republishing it. For anything public, summarise and attribute rather than reading it out.
- Sentences with three clauses fail out loud even when they read fine. Anything long should be spoken before it is recorded.
- A figure read aloud sounds more precise than it is. Speak the caveat with the number or the audio overstates the source.
Questions people ask
How is this different from text-to-speech?
Text-to-speech reads the page, including the parts that only work visually. This is a rewritten script: chart figures spoken, sources named, and sentences restructured so a listener can follow without going back.
Can I do several articles as one episode?
Yes, and it makes a better episode. A briefing with a segment per source, crediting each, is more listenable than one long reading and is easier to keep to ten minutes.
Is it all right to publish this?
Summarised and attributed, generally yes; read out in full, no, that is republishing someone's work. For internal listening the line is more relaxed, but the attribution should still be in the audio.
Bring your own web pages
Everything you save lives in one workspace, so the podcast is built from your sources — not from a model's memory of the internet.
Open Ancher →Not a web page? The same podcast also comes from YouTube video, X thread, LinkedIn post.