Back to browse
Unread, turns your unread newsletters into a daily podcast

Unread, turns your unread newsletters into a daily podcast

by benfosterdev·Mar 6, 2026·2 points·2 comments

AI Analysis

●●●BangerSolve My ProblemShip It

Honest problem-solving: converts newsletter debt into listenable daily podcasts via RSS.

Strengths
  • Solves a real, relatable pain point: newsletter subscription overload meets listening preference
  • Smart tech stack: Postmark for email, Cloudflare Durable Workflows, Supabase—production-ready choices
  • Genuine aha moment: newsletter-to-podcast is a novel angle, not a wrapper on existing tool
Weaknesses
  • Early alpha: unclear how well text-to-speech handles dense content or how personalization scales
  • Depends on user discipline: forwarding workflow might create friction vs. organic reading
Category
Target Audience

Newsletter subscribers who prefer listening to reading

Similar To

Readwise Reader · Matter · Pocket Casts

Post Description

I subscribe to about 30 newsletters to stay "on top of things". I read maybe 3.

They started to fill up my inbox so I did the responsible thing and created a rule that moved them to a dedicated folder "Newsletters". Now I don't read any.

It dawned on me that I'm far more likely to listen to a podcast than read a newsletter. What if I could take all those unread newsletters and turn them into my own personal podcast?

So I built Unread. You forward your newsletters to a unique address, and turn them into a conversational podcast episode covering everything you didn't read. You subscribe to the RSS feed in whatever podcast app you already use.

It was a fun project to build (you can get more details at https://x.com/benfosterdev). Some of the more interesting technical pieces:

- Postmark for incoming emails - Cloudflare for hosting everything - really enjoyed their durable workflow framework - Supabase for DB and Auth - React frontend - Claude for content extraction, synthesising, research and script writing - Google Gemini TTS for generating audio (nice support for SSML)

I started with ElevenLabs for TTS. It was great and pretty realistic but way too expensive. OpenAI TTS was okay, but barely. Gemini 2.5 TTS hit the quality/price point on the mark (listen for yourself) and has enough control that you can create conversational sounding content instead of two robots talking in a vacuum.

There's still a lot of refinement to be done on the script/audio side of things to make it more engaging and human-like but I'm already using it daily to get my tech updates.

Next important feature will be to schedule episode creation.

You get 5 free episode credits on sign up. Feedback appreciated.

Similar Projects

AI/ML●●Solid

Podscript – Podcast/YouTube Transcription CLI

Outputs ready-to-use Markdown with speaker diarization and timestamps, accepts Apple Podcasts/YouTube/RSS links, and can run fully locally or use ElevenLabs for higher-quality diarization. Not groundbreaking — speech-to-text pipelines already exist — but the one-command UX, RSS browsing/search flags, and explicit local-mode make it genuinely useful for folks who want tidy transcripts without wiring together multiple tools.

Solve My ProblemNiche Gem
timf34
103mo ago