WhisprStream started from a simple pain: too many X threads and articles worth finishing, and a day that never offered enough sitting-still time to read them.
I got tired of bookmarking forty threads I never opened. That sentence is almost the entire product brief. The rest is engineering and taste.
The real problem
The writing I valued most was not missing. It was stranded in the wrong interface. Threads demanded full attention. Newsletters piled up. RSS was pure but unread. Podcasts had my commute, but the open web did not.
I did not need another infinite feed. I needed the feeds I already chose — played as sound.
What I built
WhisprStream turns X threads, posts, long-form pieces, RSS items, and pasted web articles into AI-narrated audio stations. You can listen free on public stations with no account. You pay when you want to generate your own multi-source station.
Along the way the product grew the distribution surfaces that make a station feel real: shareable handles, embeds, podcast RSS, even a listen-only path for Meta Ray-Ban Display.
What I use it for
- Founder and builder threads I would otherwise half-skim
- A handful of newsletters that deserve full attention
- One-off essays that would die in a tab
I still read when reading is better. I stop pretending I will read when I will not.
Build-in-public notes
This is an early-access product. Voices improve. Limits exist because narration costs money. Public listening is intentionally free so the medium can be evaluated without a credit card. If something is rough, it is because it is real software shipping in the open — not a pitch deck.
If you are in the same spot
Start with the lowest commitment path: press play on a public station. If you hear your own backlog in that experience, sign in with X and put your actual sources into a station.
Further reading if you want the practical manuals: threads as audio, RSS as a podcast, killing the read-later graveyard.
Thanks for listening — even if you are still reading this one with your eyes.