← Blog · 10 July 2026

Habits have changed. We read less sitting in front of a screen and listen more: in the car, on public transport, while cooking, while working out. Podcasts installed the reflex; articles, meanwhile, stayed text.

The real barrier isn't technology

Converting an article to audio isn't hard: any journalist can paste their text into a text-to-speech tool, download the MP3, import it into the CMS, place it at the top of the article. It takes ten minutes.

Ten minutes per article. For a newsroom publishing ten articles a day, that's one to two hours of daily work, purely moving files around. Nobody does it — and that's perfectly rational.

What blocks it: friction, not cost

Large press groups have brought voicing in-house: they have developers, and they wired it into their publishing chain. The others — the vast majority of titles — have neither the team nor the budget to outsource a custom development.

The result: around 80% of news titles have no audio version, not because they don't want one, but because the integration cost exceeds the perceived value.

What changed in 2026

Two things. First, the quality of synthetic voices: you can no longer tell a good generated voice from a human one on a news text. Second, the cost: producing an hour of audio now costs a few tens of cents.

So the question is no longer "is it feasible?" but "does it fit into my workflow without changing anything?".

Zero friction: the only criterion that counts

That's the WeDispatch principle. The journalist writes their article, clicks Publish, and the audio is there — in seconds, before the first reader even reaches the page. No handling, no files to move, no habit to change.

Audio becomes what it should always have been: a property of the article, not a project.