Hacker News Reader

Comment by xenocratus | Hacker News Reader

I did something very similar to this, for myself, but which takes a web page for an article, cleans up the text to preserve the article only, and converts it using TTS, then adds it to a cloud bucket with an RSS stream, and I use that RSS link as a podcast in Apple Podcast.

View on HN

I do similar things, just use ElevenLabs for tts - a good selection of voices.

A good protip - tell an agent to rewrite the article slightly for better listening flow. Some subtle changes in sentence structure etc can make listening way more pleasant.

Same here.

https://blog.nawaz.org/posts/2024/Apr/reading-articles-via-p...

I use the TTS on Google Cloud. It's effectively free.

Oh man. I did something similar but didn’t do the last mile with the cloud bucket and RSS stream and that seems key. I never have the energy to manually copy the mp3s over to the phone.

I did both of that too! At the end I made my own app to listen to the episodes.

I have syncthing, I just copy a file into the shared folder.

I did something similar and just put the files on an old laptop and used Tailscale. That gives me an https URL and Apple Podcasts downloads the files fine.

I did an even simpler thing. I had it connect to gmail MCP and email it to me as an .mp3 file which I then play on my phone.

What TTS do you use? I find the OpenAI ones really unpleasant to listen to and elevenlabs is better but still not quite right somehow.

Try fish.audio - their S2 Pro has been very decent. The S2.1 Pro might be better but haven't had a chance to try it.

Same here. 75% of my podcast listening time come from content generated by claude and voxtral these days (tts from mistral).

how do you find Voxtral quality?

Very good. I have a listening experience close to elevenlabs. But sometimes you can get "failed" segments so you need to detect them and retry/repair automatically. One example is the fading volume.

Sounds great! Any interest in open sourcing your approach?

I might do, need to find some time to find a good balance between security and usability by others. It's currently running on the public web, so I had to have some funky security to prevent bots from triggering it and using my money for TTS :)

But worth noting that it took me a few hours to make, and it's all vibecoded, meaning the code is likely not that nice, and that you could probably replicate it fairly easily just from my description above.