My browser had a bookmarks folder called Read Later. It held three hundred articles.
Here’s the brutal part: I never read any of them. Reading requires a quiet block of time at a desk, and my life doesn’t contain blocks like that. It contains a forty-minute commute, ten minutes in lines, fifteen minutes of dishes.
Three months ago I started an experiment: convert those articles to audio and listen instead. Sixty articles later, that folder is emptying for the first time in years. This is the complete system.
Why listening survives and reading doesn’t
It’s not that listening is superior. It’s that commuting is dead time.
Reading takes your eyes and hands. On a train, your eyes watch the doors and your hands hold the rail. Your ears, meanwhile, are completely idle—count the earphones in any metro car and you’ll get it.
So I didn’t find extra time; I cashed in dead time. Twenty minutes a day, five days a week, adds up to one long article a day plus change—roughly two books’ worth of information a month. I had never actually run that math before.
The full workflow
Step one: choosing articles. Not everything converts well. My filter is simple: pieces built on narrative and argument work; anything heavy on tables, charts, or code snippets doesn’t. Listening costs you little comprehension on the first kind and a lot on the second.
Step two: converting. Paste the text into Soundwaver, pick a voice you can live with, generate—takes under a minute—and export to your phone. For long pieces I strip ads and introductions first, keeping just the body; a ten-minute file should not sound choppy. I name every file date-first, so anything I want to find again later sorts itself.
Step three: the queue. My phone’s playlist is the queue. New articles go in the back, finished ones get deleted. One rule only: the queue caps at twenty files. To add number twenty-one, I have to listen to the oldest one first. That single rule cured my habit of hoarding audio—the same habit that filled the Read Later folder in the first place.
What three months actually felt like
Honest version: not a smooth ride.
Week one, I caught myself drifting. When you read, you stop when tired; when you listen, you drift and the article keeps going—three paragraphs gone before you notice. Two adjustments fixed most of it. First, speed up to 1.25x; slow voices invite wandering. Second, skim the headings before listening so the structure is already in your head. Drift dropped noticeably after that.
Second discovery: re-listening has real value. I never re-read articles, but replaying audio feels effortless. Hard concepts click on the third pass in a way they never did on paper. It feels closer to radio than to reading.
One more shift: I attach listening to chores. Dishes, laundry, walking—hands busy, ears calm. Sitting still is when I drift; moving is when I focus. That trick is stolen from experienced podcast listeners, and it works.
The bonus nobody planned: English
Once the system ran smoothly, I threw English articles into the queue. Listening comprehension turns out to be a completely different muscle from reading.
Brutal at first—two or three passes per article. A month in, something opened up. Normal-speed English long reads now land around eighty percent comprehension for me. Completely unplanned, probably the biggest gain of the whole experiment.
Tip: start with slower, crisper voices and graduate to normal speed. And don’t pause to look up words—rhythm beats vocabulary; let the unknown words sail past like song lyrics.
My weekly rhythm and backup plan
Beyond the workflow, rhythm matters. Sunday night I batch-convert: the five to seven articles I want that week, generated in one sitting, then downloaded for offline playback. Weekdays, I never open a laptop—the commute just starts playing. Tunnels, business trips, flights with no signal: none of it interrupts anything.
Batching has a quieter benefit, too: judgment stays consistent. Voice picks and speed settings made in one focused session match each other far better than choices scattered across tired weekday evenings.
Two small details. One: the same voice for two straight weeks gets tiring, so I generate two versions of each file and alternate between them—change the channel when it goes stale. Two: filenames carry the speed tag (1.0x or 1.25x), because dense articles get listened to at normal speed and easy ones at fast-forward, and future-me needs to know which was which.
The time I broke my own system
For fairness, a failure. In month two I did a purge-style mega conversion: twenty-plus long articles, all at once, straight into the queue. Inventory means freedom, I reasoned. Three days later I was slacking—opening the playlist to a wall of unheard files produced exactly the same dread as those three hundred bookmarks, and I started skipping around, finishing nothing properly.
I cut the cap from twenty files to twelve after that. I’d rather convert twice on a weekend than let the queue become a second source of stress. The strangest lesson of this whole experiment is that the tool was never the problem. Hoarding was. Articles pile up; audio files pile up in exactly the same way. Every rule in this system exists to guard me from myself.
What the habit itself changed
The biggest shift wasn’t volume; it was what I choose to save. I used to bookmark on title alone. Now I ask one question first: does this piece work as audio? If not, it stays unread with my blessing. The folder stopped being a landfill and became a queue—and that mental shift outranks any tool in this story.
Second shift: I reply faster. A friend drops a link saying “read this,” and it used to sit for a week. Now it’s heard on tomorrow’s commute, and my reply comes back with actual specifics. Friends seem to think I’ve become more attentive. I just have a system.
Third: notes. Anything I finish on the train gets three lines written at my desk within minutes of arriving. Listened-and-unrecorded equals never listened at all. Those two minutes are the last step of the workflow, and they’re not optional.
What not to convert
Three types. Tutorials—steps and screenshots are the point; audio turns them into fog. Listicles—ten points means you’ve forgotten point three by point seven. Very short news items—four hundred words spends more time in conversion than in listening.
One principle covers it: if a piece’s value lives in its argument, listen; if it lives in its visuals, read. The check takes five seconds.
If you want to start
Copy the workflow: pick one long-saved article, paste it into a text-to-speech tool, export it, listen tomorrow on your commute. For tool choice, I compared the popular free options in my free TTS tools guide—licensing, language quality, monthly limits.
Give yourself a week of grace. Drifting early is normal, like getting winded on your first jog; ears need training too. And if you’re curious why synthetic voices got this natural, what TTS actually is covers the mechanics—it genuinely helps you pick voices.
That Read Later folder? Down from three hundred to under fifty. Not because I deleted anything—because I finally heard them.
