Build a Private Commute Podcast from Your Reading List

A 30-minute commute is one of the few stretches of the day that nobody else is asking for. Most people fill it with a Spotify playlist or whatever podcast they have been chipping away at for two weeks. There is a third option that almost nobody sets up: play the articles you actually meant to read today, narrated by an AI host, in your normal podcast app.

This article is for knowledge workers who keep saving links and never opening them. The setup is one Show, a few RSS feeds, and a schedule that matches your morning. By the time you get off the train (or out of the car, or off the bike), you have skimmed the day's reading list by ear and know which two or three pieces are worth a real read at your desk.

Why commute time and your reading list never meet

The reading pile and the commute happen at different times of day. Reading happens (in theory) when you have a screen, focus, and a free hour. Commuting happens with a backpack on, one hand on a handlebar, or eyes fixed on traffic. The pile grows in the first window. The commute happens in the second one. Audio is the only format that fits the second window, and most of what is available is either music or somebody else's podcast.

There used to be a soft bridge here in the read-later category. Mozilla retired Pocket in 2025, so the most familiar way to save articles for later (with read-aloud as a bonus) is gone. Feedly and Instapaper still offer text-to-speech inside their reader apps, but you are listening one article at a time, in the reader, with no podcast app, no automatic delivery, and no summarization. That works for the one long essay you actually want voiced. It does not turn into a habit.

The gap is a daily 15-minute audio digest of the things you already follow, sitting in the same podcast app as everything else you listen to.

What "private commute podcast" means

A private commute podcast is a podcast feed that nobody else subscribes to. The feed URL is unlisted, your podcast app fetches it on its own schedule, and a new episode shows up the way episodes from any other show do. What distinguishes it from a public podcast: the episodes are generated automatically from RSS feeds you choose; the schedule is yours (daily before 7 AM, weekly on Monday, whatever fits the commute you actually have); and the audience is you, plus anyone you share the URL with directly.

Existing options

This category has some history. ListenLater is a long-running service in the same shape: you send it an article (by email, by forwarding a newsletter, by RSS), and it gives you back a private podcast feed that pronounces the article when your podcast app fetches it. If your habit is "I find a piece during the day and want to hear it on the way home," ListenLater is built around that motion.

Lisnify works the other way around. You hand it a small set of RSS feeds up front, pick a schedule, and the system decides each morning which articles from those feeds belong in today's episode. Multi-host conversation rather than single-voice readout, summarization rather than full-text narration, and a fixed delivery time. Both approaches are reasonable. They optimise for different ends of the commute reading problem: one for "voice this thing I just found," the other for "hand me a 15-minute brief at 6:45 AM every weekday."

Pick the feeds that earn the slot

A commute episode is worth roughly 10 to 15 minutes. Lisnify caps each episode at 10 articles, and in practice 5 to 8 is where most commute Shows settle. That means each feed you add has to earn its place against everything else competing for those minutes. A few categories worth thinking about:

News, filtered. A general news fire hose is too much. A narrower slice is usually better: a Google News keyword search RSS for one topic you actually track ("EU AI act," "supply chain semiconductors," your industry), or one or two trusted publication feeds. Google News exposes any keyword search as RSS by appending the search to https://news.google.com/rss/search?q=...&hl=en-US, which is the closest thing to a "tell me when this topic moves" feed for sites that do not give you their own.

Industry or role-specific writing. This is where most of the value tends to live: the newsletter for your function, two or three company engineering or research blogs you respect, the weekly roundup somebody in your space puts out. Substack hosts a lot of these now, and any Substack URL plus /feed is the RSS endpoint.

Tech and developer signal, if it is part of your job. Hacker News and developer-blog feeds work well in audio because the entries are short and the audio version naturally turns into "skim through interesting links." If you want a daily HN slice on your commute specifically, the Listen to Hacker News as a daily AI podcast walkthrough goes into which HN feed to pick.

Long-form thinking. One or two essayists you actually read all the way through. These are the pieces audio handles best, because the value is the argument, not a chart you would have scrolled past anyway.

What does not earn the slot: catch-all feeds you subscribed to two years ago and never read, sites whose value is a chart or table, anything whose RSS only ships the headline (the audio summary will be threadbare), and anything you would normally read at your desk because you stop and re-read paragraphs. Audio is for triage, not careful reading.

Pick a schedule that matches your morning

Daily makes sense if your commute is a daily thing and the feeds you picked actually move daily. Weekly is the better default if your commute is two or three days a week, or if the feeds are weekly newsletters anyway (a daily brief from feeds that only publish on Tuesdays produces six empty episodes).

The generation time decides what is in the episode. If you leave the house at 7:30 AM, scheduling generation for 6 AM in your local time zone gives the system enough headroom to fetch, summarize, and synthesize before the episode downloads to your phone. The default schedule timezone is UTC, so set your local zone explicitly when you create the Show. If a morning episode is not waiting on your phone when you walk out, opening the podcast app once before you leave forces a refresh.

Pick hosts and language

In the Host tab you choose AI host voices, the language they speak in, and the speaking style. Two-host conversational works well for commute listening because the rhythm of two people trading off keeps a 12-minute episode from feeling flat. Single-host narration sits closer to a news read; some people prefer it for driving because it requires less attention.

Cross-language pairs deserve a mention. Lisnify supports reading source feeds in one language and delivering the episode in another. A common case for non-native English speakers: follow English-language industry sources, hear the daily brief in your own language. The reverse also works for language learners who want input in their target language.

The first two weeks: tune by listening

Most people set the Show up once and then have to adjust it after they actually listen to a few episodes on a real commute. Some patterns that come up:

The episode is two minutes longer than your commute. Drop the per-episode article count from 8 to 6, or shorten the per-story instructions in the Script Settings tab.
One feed is dominating. A high-volume news source can crowd out the slower newsletters you actually wanted to hear. Either drop the noisy feed to its own Show or add a script instruction to favour variety across sources.
The hosts cover too much ground per article. If you find yourself thinking "I get it, move on" three minutes in, ask the script for shorter per-article treatments and a faster pace.
The schedule is wrong. The episode lands at 7 AM but you do not put headphones on until 8:15. Pulling generation earlier (or moving to weekly) usually solves it.

These adjustments are obvious during the first week or two and almost invisible after a month. By that point the episode feels like a fact of the morning, and you stop noticing what could be better.

What commute audio is good at, and what it isn't

Audio on a commute is good at triage (figuring out what the day's reading list contains so you can decide what to open later) and at the kind of pieces where the value is the argument itself: essays, interviews, opinion writing. It is also a real habit that survives, because a 12-minute episode you finish daily beats a tab pile you never get to.

Where it falls short is harder to dodge. Code samples, benchmark tables, and screenshots of a UI are not really there in audio. The episode tells you that "a new release of X benchmarks faster than Y," but the actual numbers are in a table you cannot see. The audio version of those pieces is "this exists, here is the gist." Treat it as a flag to open the original at your desk.

Time-sensitive things are the other gap. An episode generated at 6 AM is fixed for the day. If a story breaks at 11, your commute home does not know about it. For breaking news your existing feed reader still has a job; the commute podcast is for the slower-moving reading list.

If you are about to use something from the audio in a serious context (citing a number in a meeting, acting on a recommendation), open the original. Audio summaries help you decide what to read, not what to quote.

Pair with reading, don't replace it

Once it settles in, the commute podcast becomes the first pass over the day's reading list, not a substitute for it. The morning ride covers eight pieces. Two sound interesting enough to open at the desk. The other six were good to know about and would have stayed unread anyway.

That changes the relationship to the pile in a small but real way. Instead of "I have 47 unread items and I should feel guilty," it becomes "I heard the rundown this morning, and there are two I want to read carefully." The reader app still does the actual reading. The commute podcast just decides which links earn the click.

Frequently asked questions

Won't I miss important details if I only listen?

For triage, audio is enough. For anything you would act on (a number you would quote, a recommendation you would follow, a technical claim you would apply), open the original. The episode is designed to tell you what is worth reading carefully, not to replace careful reading.

What if my commute is shorter than the episode?

A 20-minute episode on a 15-minute commute is fine; podcast apps remember where you stopped. If the gap is bigger (a 6-minute walk to the train), reduce the per-episode article count in the Sources tab so an episode runs closer to your actual commute length. Most apps also support 1.25x or 1.5x playback, which buys you back a few minutes without sounding chipmunky.

I take public transit and lose signal in tunnels. Does this work offline?

Yes, if your podcast app is set to auto-download new episodes. Apple Podcasts, Pocket Casts, and Overcast all download new episodes over Wi-Fi by default once you subscribe. The episode is cached on your phone before you leave, so a tunnel does not break playback. The same is true for flights, basement gyms, and patchy commuter rail. Note that Spotify does not let you subscribe to arbitrary RSS URLs, so a private commute feed will not show up there; if Spotify is your usual podcast app, plan to use a second one for this.

Yes, within limits. The feed URL is unlisted (built around a UUIDv7 identifier so nobody guesses it), and the Lisnify Terms of Service allow private sharing with a small number of people you actually know. Posting the URL anywhere publicly is not allowed. One thing to know: the feed XML contains your email address as the show owner's contact, because the iTunes Podcast namespace requires it. Anyone you share with can see that address.

I commute on a bike or walk. Is that different from driving?

A bit. Audio you listen to with one ear out (for traffic, bells, sirens) does not work as well at fast pace or with dense back-and-forth. A single-host narrator at normal speed is easier to follow on a bike than two hosts trading short lines. Cycling commuters also tend to prefer slightly shorter episodes because the trip itself is more attention-heavy. The fix is not the feeds; it is the host style and length, both of which you can adjust without changing what you subscribe to.

Where to go from here

Pick three or four feeds that you genuinely care about and would read if you had time, set up one Show, schedule it for an hour before you leave the house, and subscribe in your usual podcast app. The first week is for finding out what does not fit, and after two weeks the episode starts to feel like part of the morning rather than something you have to remember to use.

The step-by-step RSS-to-podcast guide covers the canonical setup for any feed in more depth. If your reading list is heavy on developer writing, the dev.to and Substack walkthrough is the focused version. If you are still figuring out which AI podcast tool fits your job at all, the NotebookLM alternatives compared by job write-up walks through the public versus private split.