Reading PDFs Aloud Without Losing Your Place

Making an iPhone speak a PDF is a solved problem — iOS will read any screen aloud, and a dozen apps will do it with nicer buttons. What's not solved, in most of them, is what you actually hear and whether you can find your place afterwards. Those two details decide whether PDF audio is useful or a gimmick.

First problem: what gets read

A PDF's raw text is not its prose. It's prose interleaved with running headers, page numbers, footnote fragments, and — in two-column documents — lines from adjacent columns spliced together. Feed that to a voice and you get the full absurdist experience: "...the results suggest — Journal of Applied Psychology — page 247 — a significant effect..."

Velo restructures the PDF on import, before any mode touches it: headers and page numbers stripped, columns resequenced into reading order, footnotes extracted and linked rather than read mid-sentence, chapters detected into a navigable table of contents. Scanned documents pass through on-device OCR. The voice reads what the author wrote — nothing else. It's the same pipeline that makes reading academic PDFs tolerable, and audio benefits from it even more than the page does, because your ears can't skip clutter the way your eyes can.

Second problem: following along

That tracking changes what read-aloud is for. Instead of audio as a background murmur, you get genuinely bimodal reading — hearing and seeing each word at once — which earns its keep on dense material, on tired evenings, and any time attention needs the extra channel. And because the audio advances the same reading position that Velo's visual modes share, switching is free: listen through the commute, open to the page at your desk, and you're exactly where the voice stopped. One bookmark, not an audio bookmark and a text bookmark drifting apart.

There's a sleep timer, and it respects the same position — tomorrow starts where the audio actually stopped.

Third problem: where it works

Velo's read-aloud runs on-device, not streamed from a cloud voice server. Cloud voices sound somewhat better — that's a real trade — but on-device means listening works on a plane, underground, and offline generally, exactly like the rest of the library, which is stored locally and readable with no connection.

What's free and what isn't

PDF import — including the full restructuring pipeline — and both visual reading modes are free, with up to 3 imported documents, so you can verify the thing that matters most: whether your PDFs come through as clean prose. Read-aloud itself is part of Velo Pro ($69.99/year or $8.99/month after a 7-day trial), alongside cloud sync across iPhone, iPad, and Mac.

Worth knowing before you buy anything, anywhere: this is text-to-speech, not narration, and no TTS app makes a synthetic voice into an audiobook performance. What it makes is every PDF you own listenable — the reports, the papers, the manuals that will never have an audio edition — with the text tracking along so the words actually stick.

← All guides