If you've tried Chrome's Read Aloud, Speechify, or NaturalReader on read.amazon.com, you already know: nothing happens, or it reads gibberish. It's not your browser. It's how Amazon renders the book. Here's the technical reason and how to actually listen to any Kindle book in Chrome.
2-minute setup. No account, no credit card.
Open any book on read.amazon.com, right-click a paragraph, and pick Read Aloud. Nothing happens. Highlight the text and try to copy it — you get a blank string, or a mess of mismatched letters. Here's why every TTS tool you've tried has failed:
Tested on read.amazon.com with a standard e-book in Chrome 140 (April 2026).
Reads the obfuscated glyph map. You hear random syllables or silence. Often skips entire pages.
Highlights the page but plays back gibberish strings like 'xgpz qhr zpm'. Their support thread has acknowledged the issue since 2022 with no fix.
Refuses to start playback on read.amazon.com — the extension detects that selected text is unreadable and silently aborts.
Same root cause. Reads the obfuscated layer. Only works if you copy text out, which Kindle blocks.
Reads any Kindle Cloud Reader book aloud with word-level highlighting. Uses OCR on the rendered image instead of relying on HTML text — bypassing the obfuscated font layer entirely.
We stopped trying to read the HTML. Here's what we do instead:
Kindle Cloud Reader builds each page as a bitmap before showing it. We hook into Amazon's rendering pipeline and capture that bitmap as it's produced.
The bitmap runs through Tesseract OCR (a WebAssembly port of the same OCR engine Google uses internally). This extracts real, readable text — the letters your eyes see — without touching the encrypted font lookup.
The extracted text feeds into our TTS pipeline. Audio plays with paragraph highlighting and word-level sync, just like reading a normal webpage. Everything runs locally in your browser — your book never leaves your device.
Free on Chrome, Edge, and Firefox. No account, no credit card.
Log in with your normal Amazon account. Open any book you own.
The first OCR pass runs in ~3 seconds. Audio starts. Page turns happen automatically as reading progresses.
Because Amazon renders Kindle Cloud Reader pages as images with obfuscated font glyph tables, not as HTML text. Chrome's Read Aloud, Edge Read Aloud, and every other HTML-based TTS tool read the obfuscated characters and produce gibberish. The only way to get real text is OCR on the rendered page, which CastReader does automatically.
No. Speechify reads the HTML text layer, which on read.amazon.com is encoded with a DRM font lookup. It will appear to work — showing highlighting — but the audio output is nonsense. This has been an open issue in Speechify's forums since 2022.
Yes. You own the book and you're reading it aloud for personal use in a browser on your own machine. CastReader runs entirely locally — no upload, no cloud OCR, nothing leaves your browser. Amazon's DRM is about stopping extraction and redistribution, not stopping you from hearing a book you own.
The browser extension only covers read.amazon.com. For the Kindle desktop app, Apple Books, and other native Mac apps, install CastReader for Mac — it uses macOS accessibility APIs to read aloud from Kindle Desktop, Apple Books, Preview, Pages, Notes, Word, Obsidian, and Notion, with the same word-level highlights as the browser extension.
The first page takes ~3 seconds to OCR. After that, pages pre-process in the background while you listen, so you rarely notice. The OCR engine runs in WebAssembly entirely on your machine — no network round-trip.
40+ natural AI voices across English, Spanish, French, German, Japanese, Chinese, and more. The extension auto-detects the book's language from the OCR output.
After playback starts in Chrome, click Send to Phone. Audio streams to your phone via Telegram — no mobile app, no sync setup.
CastReader is 100% free with no daily limits, no account required, and no gated voices. Unlike Speechify's $139/year tier or NaturalReader's paywall.
Install once. Open any Kindle Cloud Reader book. Click ▶. Listen.
Canvas rendering defeats every TTS tool — here's the fix
PDFs that Chrome, Speechify, NaturalReader can't read
Canvas-rendered Chinese e-books, read aloud in browser
The full Kindle support page with compatibility details