The More Beautiful World Our Hearts Know Is Possible
I was excited to listen to the Audible edition of this book, narrated by Paul Geiger, but I found the narration difficult to listen to. Since many other readers (listeners) had a similar experience, I am hoping the present page will be useful!
I decided to try a text-to-speech (TTS) tool instead. Charles publishes his books in full on his website archive, one chapter per page, under a Creative Commons license. The current page simply stitches the into a single document, optionally removing the endnote sections, to make it easy to load into a TTS reader.
Downloads
There are two variants, with three formats each. The endnote-free version is what I recommend for text-to-speech, as endnote citations (while interesting in many cases) break the flow.
TTS-friendly (no endnotes): EPUB, HTML, Markdown.
Complete (with endnotes): EPUB, HTML, Markdown.
Text-to-speech tools
If you haven't used a modern neural TTS tool before, my experience is that the quality (as of July 2026) is now on par with a decent human audiobook narrator, apart from occasional audio artifacts or hiccups when the AI "gets confused". (Sometimes hilariously so! But even with the occasional "uh, um" and hesitant tone, I still much preferred it to the Audible version.)
I listened to the book with ElevenReader, a free iOS / Android / web app from ElevenLabs. It was easy to import the EPUB: just download the app, download the EPUB file (or import the link from this page), choose a voice, and you're rolling. I found it had the usual playback features I was hoping for: play, pause, rewind, set speed, and a bookmark feature with the option to write notes to myself. I was going to try Speechify but didn't. I bet you can find plenty more with a web search! Below are some options and descriptions that Claude generated; take with a grain of salt...
Free / open source
- Kokoro — Apache-licensed open-weights TTS model (~82M parameters). Voice quality is startlingly close to commercial options. Runs locally, Python-friendly.
- Piper — fast, fully offline neural TTS; runs on essentially anything, including a Raspberry Pi. Actively maintained under the Open Home Foundation.
- edge-tts — Python CLI that streams Microsoft's neural voices for free. Very natural; relies on an undocumented endpoint, so it can break.
- Calibre — the "Read aloud" feature in Calibre's viewer uses your OS's TTS by default (quality depends on your OS — decent on macOS, good with a Piper backend on Linux). Zero setup if you already read EPUBs in Calibre.
On your phone, no installation
- Apple Books (iOS) and Google Play Books (Android) both read EPUBs aloud for free using on-device neural voices.
Attribution and license
The text of The More Beautiful World Our Hearts Know Is Possible is © Charles Eisenstein and licensed under a Creative Commons Attribution 4.0 International License. The files above inherit the same license. The changes from the original are minimal: chapters are combined into a single file (and in one variant, endnote sections are removed).