ThreadCast
Listen to Reddit like a podcast
Hey IH 👋
Solo founder here (Pixel Labs). A couple of weeks ago I posted that I was building ThreadCast. It's live on Google Play now, so here's the launch and the one decision I'm still chewing on.
The itch: I read a lot of Reddit, but always on a tiny screen at the worst times, on a commute, at the gym, doing dishes. I kept wanting to just listen. Every text-to-speech app I tried read the whole thing in one flat robot voice. A 40-comment thread turned into an unlistenable monotone wall. You lose who's saying what almost immediately.
So I built ThreadCast. It takes a thread (post + top comments) and reads it aloud like a podcast, and the part I'm proudest of: every commenter gets their own voice. Drop into a thread, hit play, and it actually sounds like a conversation instead of one person droning. That one detail changed how listenable a long thread is more than anything else I tried.
The technical bet that made it hard (and, I think, worth it):
The AI neural voices run 100% on the device. No cloud, no account, nothing leaves your phone. Most read-aloud tools ship your text to a server to synthesize; ThreadCast deliberately doesn't. The upside beyond privacy is that once you've downloaded the voices and a thread, it all works offline, so you can listen on a plane with no signal. Getting good neural voices to run locally on a phone was the bulk of the engineering, but it's the whole thesis, so I wasn't going to compromise on it.
What it actually does today:
· Multi-voice playback, up to 11 on-device AI voices, a different one per author.
· Comment control: it reads the post + top comments, and you choose which comments and how deep, so you're not stuck listening to 200 replies.
· Queue + local save: line up posts, and threads (post and comments) get saved locally for offline listening later.
· Sleep mode: 10 ambient sounds + a fade-out timer, because I mostly use this in bed.
· Lock-screen, Bluetooth, and media controls, no account needed.
· Full TalkBack support: an app that reads Reddit aloud should be genuinely great for people who need it read aloud, so accessibility wasn't an afterthought. Every control is labeled and swipe actions are exposed as screen-reader actions.
· MP3 export (Premium): save any thread as a shareable audio file and listen in any podcast app.
Free vs Premium: free gets you your phone's system voices, browsing, the queue, comment control, sleep mode, and lock-screen controls, basically the whole listening loop. Premium is the on-device AI voices + MP3 export. No ads, no tracking, no VC, just me trying to build something people pay a fair price for.
There's also a free Chrome extension for desktop, and it's not a lite version, it's the same core: four on-device voice engines, the multi-voice narration, comment selection, a docked side panel, and sleep mode. The only things still coming to the extension are MP3 export and reading stuff beyond Reddit (PDFs, articles, your own text). iOS is later.
The thing I keep going back and forth on, and where I'd genuinely love this community's take:
1. How do you price a local-first app? My whole moat is that the AI runs on-device, which means no per-use server cost, but also no usage meter to anchor a subscription to. One-time vs subscription for something with basically zero marginal cost has me stuck.
2. The per-commenter-voices thing: does that actually sound like a real feature to you, or like a gimmick? It's the part I bet the product on, and I'm curious whether it lands the same way for people who aren't me.
Play Store: https://play.google.com/store/apps/details?id=app.threadcast
Chrome extension: https://chromewebstore.google.com/detail/threadcast/dhbcjkjfhebbjffojpcmfoffhchijpkn
Site: https://threadcast.app
Happy to get into any of the on-device/offline details if that's interesting. Thanks for reading.
Hey IH 👋 Solo founder here, building under the name Pixel Labs.
I kept wanting to listen to Reddit on my commute instead of reading on a tiny screen — so I built ThreadCast. It turns a thread (post + top comments) into a podcast, with a different AI voice per author so it plays like a conversation, not one flat monotone reader.
Privacy is the whole thesis: the AI voices run 100% on-device. No cloud, no account, nothing leaves your phone, works offline. Most read-aloud tools ship your text to a server — ThreadCast deliberately doesn't. That's a harder engineering path, but it's the point.
Where it's headed: MP3 export — save any thread as a shareable audio file you can play in any podcast app, send to a friend, or keep for a flight. That lands with the Android app.
Status: Chrome extension live and free today; Android app + MP3 export landing soon. iOS later.
Monetization (the IH question 🙂): free with your phone's system voices + all the core listening; the plan is to charge for the on-device AI voices + MP3 export once the Android app ships — indie pricing, no VC.
Would love feedback from this crowd — on the on-device/privacy approach, the voice quality, and how you'd price a local-first app. Happy to share what I learn.
-
2
This is a good idea, but I’d be careful not to position it only as “Reddit to podcast.”
That sounds fun, but maybe a bit novelty-app-like.
The stronger angle is probably: private, offline listening for long Reddit threads.
That makes the on-device part feel like a real reason to choose ThreadCast, not just an engineering detail. The user is not only buying voices or MP3 export. They’re buying a cleaner way to save, listen, and revisit high-signal threads without sending the content anywhere.
For pricing, I’d probably not charge only for “AI voices.” I’d bundle the paid version around the full listening workflow: better voices, MP3 export, offline library, and maybe saved thread collections.
Happy to put a tighter version in writing if useful. I’d map the free vs paid split, Android launch pricing, and which user segment is most likely to pay first.
-
1
This is genuinely helpful, thank you. "Private, offline listening for long Reddit threads" is tighter than how I've been pitching it, the "long threads" part especially, since that's the moment a screen actually gets painful.
You're also right that bundling the paid tier around the whole listening workflow beats selling it as just "better voices." MP3 export is next, and an offline library plus saved thread collections are on my list, so framing those together as the paid experience makes far more sense than leading with the voices alone.
This is my first side project, so I'm still working out the pricing story, and an outside read like this is exactly what I needed.
And yes, I'd genuinely take you up on the tighter written version if you have the time. No pressure at all if not.
Really appreciate you taking the time to think this hard about a stranger's launch. That kind of generosity sticks with you when you're just starting out.
-
2
Glad it helped. And yes, this is exactly the kind of thing where a tighter written version would be more useful than another scattered comment.
The main question is not just “what should cost money?” It’s what paid version feels natural for someone who already wants to save and listen to long threads without friction.
Drop your email and I’ll send over a compact version. I’d keep it focused on the free vs paid split, Android launch pricing, and the first user segment most likely to actually pay.
-
1
That reframe is the bit that clicked, it's less about what to fence off and more about what someone who already saves and revisits long threads would pay to make smoother.
Where I'm landing: free is the frictionless loop (queue a thread, listen now), paid is the keeping part you come back to, downloaded and organized so it lives on your device.
So you pay to keep and revisit, not to unlock listening. And yes please to the compact version, especially the 'which segment pays first' part. Would be really appreciated, thanks again Aryan.
For my email, you can reach out on contact@threadcast.app
-
1
Just sent you a note.
Kept it focused on the free vs paid split, Android launch pricing, and which ThreadCast user is most likely to pay first.
-
1
Appreciate you taking the time, Aryan.
Even just the back and forth here gave me a lot to work with, so thanks again for engaging with a random solo founder's launch.
-
-
-
-
-
2 Comments
Wow! This is really cool and I like the idea. It’s very entertaining but I think you could add a bit more emotion because it sounds like they are just robots talking. By making the AI bot speak with more emotion would sound a lot better and be more entertaining.
Really nice website though!
Thanks, really appreciate you actually giving it a listen 🙏
That's fair, and it's the tradeoff I keep wrestling with. The voices run fully on-device so nothing ever leaves your phone, which keeps it private and works offline, but it also means I can't lean on the big cloud models that nail emotion. That expressiveness is the one thing cloud TTS still wins on.
The app actually has a few voice tiers, and the top one (Studio) sounds noticeably more natural than the lighter ones, worth a listen in there.
More emotional, expressive voices are high on my list though. On-device TTS is moving fast and the gap closes every few months.
And thanks for the kind words on the site, that made my day 🙂