How long does it take to add audio to a blog of 300 existing articles?
Catching up an existing blog comes down to two numbers people keep confusing: machine time, counted in minutes, and monthly volume, which actually sets the pace. A full costed calculation on 300 articles, no flattering rounding.
The question comes up the moment an editor considers audio for a blog that already exists: "I have three hundred articles live, I'm not starting from scratch. How long to voice them all, and what does it cost?" It deserves a real calculation, not a promise. Here is the one for three hundred articles, worked end to end.
Two durations people mix up, and you have to separate them
The first thing to understand is that there are two different times behind the question, and you answer wrong when you blend them.
The first is machine time: how long the engine takes to produce one article's audio. For a typical blog post, say eight hundred words, generation is a matter of seconds, not minutes. Run as a batch, a queue of three hundred articles therefore does not stretch over days: it processes in an order of magnitude that runs from a few minutes to about an hour of work, in the background, with nobody sitting in front of the screen.
The second is human time, and this is where the real difference with a voice recording shows up. You do not re-read three hundred times, you do not re-record, you do not edit. You point at the corpus once, and the tool walks the queue on its own. The human time of catching up three hundred articles is measured in minutes of setup, not weeks of production.
The scary figure, "three hundred articles," is not the right scarecrow. What actually sets the pace is something else.
What really paces the catch-up: your monthly volume
A volume-based service does not bill per second of compute, it bills a number of articles voiced per month. Three hundred articles at once means three hundred articles consumed. The real planning question is therefore not "is the engine fast" (it is), it is "does this catch-up fit inside the volume I take that month."
Two paths, and both are honest. Either you take, in the catch-up month, a volume that covers the three hundred archive articles on top of your current output, and everything is voiced in one pass. Or you spread it: one hundred articles a month for three months, starting with the ones that matter. The first path is fastest, the second smooths the spend. Neither depends on the power of the engine, only on how you want to spread the cost.
The cost, in figures, without dressing it up
Let's bring this back to money. At the scale of a blog, price is reasoned per article voiced, and it drops as volume rises: the more you take in a month, the less each unit costs. The order of magnitude, for a catch-up of three hundred articles, is around one pound or euro per article, often less at volume, so a few hundred for the whole archive. The exact grid, tier by tier, is on the pricing page: we don't print a figure here that could go stale, because a wrong number on an invoice is worse than a missing one.
This amount is a one-off. Once the archive is caught up, you only pay for current output, article by article, at publication. The catch-up is a spike, not a rent.
And it should be compared with the only alternative that gives the same steady reading quality: having three hundred articles read by a human voice. There the cost runs to tens per unit and the delay to weeks of studio time, as we lay out in how much article audio costs. It is not a close call.
Where to start: the order that pays off immediately
Three hundred articles are not equal. A well run catch-up does not process them in an arbitrary order, it starts with the ones that pay.
First your most-read articles. They already draw traffic, so their audio version will be listened to tomorrow. Voicing the top of your audience first means banking the benefit before the catch-up is even finished. Then your evergreen content, the reference pieces read all year that will keep pulling visitors. Last, the low-traffic old archive, which has the least to gain from jumping to the front of the queue.
This order has another merit: it lets you judge on the evidence after the first fifty. If the result suits you on your flagship pieces, you continue; if not, you have committed only a fraction of the volume.
How you point at three hundred articles without typing them one by one
That leaves the mechanics: how to point the tool at an existing corpus without pasting three hundred URLs by hand. This is exactly where integration matters, and it is the question every WordPress editor asks. Our WordPress plugin discovers your published articles and voices the queue without you touching the theme; for a custom CMS, RSS feed discovery does the same job from your feed address, and the API covers bespoke cases. In all three, you describe the corpus once, and the queue then processes itself.
The calculation, in one sentence
Voicing three hundred already-published articles is not days of processing nor weeks of work: it is on the order of an hour of background generation, a few minutes of setup, a one-off spend of a few hundred that depends on your volume, and an order of play that starts with your most-read articles so the benefit lands before the end. The real trade-off is not technical, it is budgetary: all at once, or spread over three months.
Give your articles a voice with WeDispatch
This blog is itself voiced by WeDispatch. Curious how it sounds on your content?
Book a demo