What Speed to Listen to Audiobooks: What the Studies Show

Start at 1.0x. If the book feels easy to follow, move up in small steps, 1.1x or 1.25x, and stop at the last speed where you still catch who is speaking and why. Treat anything past 2.0x as a setting for material you can afford to half-miss. This is a choice of setting, not a verdict on how anyone should listen, and the studies below only reach part of the question.

What the comprehension studies tested

Every study here used a named group of people, so each one comes with its sample and its design. The evidence starts in 1968, with Barabasz in the Journal of Communication (PubMed record). He used 118 college students who heard two course lectures, one at normal speed and one compressed to about 1.5x, and tested them on recall right away and again two to three weeks later. He found no significant difference between the normal and the faster groups.

A 2018 study by Song and colleagues gave a less comfortable result. It was a randomized pilot with 54 medical students at the University of Kentucky, who watched two short medical videos at 1.0x and at 1.5x and took a test straight afterward. On the test about ultrasound artifacts, the 1.5x group scored lower, 61.4 versus 72.7. On the transducers test the gap was smaller, 66.9 versus 73.8, and it was not statistically significant. So, even at 1.5x, new and unfamiliar facts in a short video cost these students something on one of two tests.

Studies that go past 1.5x

Some studies push the speed higher, and they point the same way: the cost is small up to about 2.0x and grows after that.

• Murphy and colleagues (2022). This experiment had 231 UCLA undergraduates, after exclusions, who watched lecture videos at 1.0x, 1.5x, 2.0x or 2.5x. They took a test right away and another one week later. There was little cost at 1.5x or 2.0x, and the 2.5x group did worse than the 1.0x group. A second experiment found that watching a lecture twice at 2.0x had no advantage over watching it once at 1.0x. The author-provided full text has the detail.
• Chen and colleagues (2024). In Educational Psychology Review, 200 UCLA undergraduates heard audio-only science clips at speeds from 1.0x to 2.5x, then answered 30-question tests right away. Speed had no statistically significant effect, though scores fell about 6% from 1.0x to 2.5x. The authors advise caution past 2.0x.
• Ritzhaupt and Barron (2008). This experiment used 305 participants and narration at 1.0x, 1.5x, 2.0x and 2.5x. Cued recall and recognition were significantly lower at 2.5x.
• Kuperman and colleagues (2021). The study compared matched reading and listening passages. Listening comprehension held at 270 words per minute and dropped steeply at 315.

The Kuperman numbers are words per minute, not a speed setting on an app, so they do not convert neatly into 1.5x or 2.0x. The useful part is the shape: comprehension held at 270, then dropped steeply at 315.

What none of these studies tested

Each of these studies used lectures, educational video or science clips, with short immediate tests or a test one week later. None used a commercial novel or a full-length audiobook. So, none of this tests fiction, and how well a story holds at speed is for you to judge by trying it.

One more study is worth a mention for its limit. Buchweitz and colleagues scanned nine listening participants with fMRI and compared familiar with unfamiliar expository topics. That is a comparison of how familiar the subject was, not of fiction against nonfiction, and it does not give a speed to pick.

Here is a test you can run yourself. Play one chapter at the new speed. When it ends, ask whether you can still follow who is speaking and why they are doing what they are doing. If you can, stay. If you lose a name or a plot turn, go back one step, say from 1.5x to 1.25x, and try the next chapter.

What narrators and Audible say

These are opinions from named people, not findings. Per NPR, January 22, 2025, Phillipa Soo said the best speed is the one the book was recorded at, and Jeff Daniels and Glen Powell agreed. Narrator Nan McNamara, who told NPR she had recorded nearly 350 audiobooks, said speeding up can cost some of the performance. She also said it is worth it if it is the difference between listening and not listening. Audible's head of brand and content marketing, James Finn, said the company wanted to start a conversation, not prescribe one setting.

Those three views do not agree with each other, and they do not need to. One is about hearing the performance as recorded; another is about getting through a book at all. Which one matters more is your call.

What Audible reported about its own customers

Audible published figures on January 15, 2025, in a newsroom post. They describe Audible customers only. They say nothing about listeners on other apps, or about audiobook listeners in general.

• 1.0x remained the most popular speed.
• About 7% of customers had tried 1.5x or faster.
• After 1.0x, the next most common speeds were 1.2x and 1.5x. NPR's reporting of the same data lists 1.0x, 1.2x, 1.5x, 1.7x and 2.0x in that order.
• Nonfiction was as likely as fiction to be heard faster. Romance and erotica were most often heard at 1.5x, with business and careers close behind at 1.4x.

That "tried" in the 7% figure matters. It counts customers who tried a faster speed, not customers who stay there.

How many hours each speed gives back

The arithmetic is simple: listening time equals the runtime divided by the speed. Here it is for a 10-hour book and a 30-hour book.

For a 10-hour book:

• 1.0x takes 10 hours.
• 1.25x takes 8 hours.
• 1.5x takes 6 hours 40 minutes.
• 1.75x takes 5 hours 43 minutes.
• 2.0x takes 5 hours.
• 2.5x takes 4 hours.
• 3.0x takes 3 hours 20 minutes.

For a 30-hour book:

• 1.0x takes 30 hours.
• 1.25x takes 24 hours.
• 1.5x takes 20 hours.
• 1.75x takes 17 hours 9 minutes.
• 2.0x takes 15 hours.
• 2.5x takes 12 hours.
• 3.0x takes 10 hours.

Pausing, rewinding and replaying a chapter add time back, so treat these as the best case. The long book shows the point of the exercise most clearly. Going from 1.0x to 1.25x on a 30-hour book saves 6 hours, which is a bigger gain than going from 1.5x to 1.75x saves on a 10-hour one. A small step on a long book buys a lot of time. For a long book where this matters, the runtime example in the 11/22/63 audiobook length page shows the math applied to a real title.

Where to set it

Every one of these apps has a speed control. What differs is the range and how it saves.

• Audible allows 0.5x to 3.5x in 0.1x steps, and it saves a speed for each title. The setting is described on Audible's narration speed page.
• Libby offers presets plus a slider from 0.6x to 3x, per its help page.
• Apple Books uses a dial that you drag from the 1x control, as Apple's support page describes it.
• Libro.fm lets you choose a speed and save a default, per its support page.
• Spotify has a speed control on mobile and web, as its audiobook support page explains.

If you play DRM-free audiobook files, two apps add extras. Voice Audiobook Player offers skip silence and a louder volume setting. On Tape offers skip silence too.

Common questions

What speed do most people listen to Audible?

Audible's own figures from January 15, 2025, cover Audible customers only. They say 1.0x remained the most popular speed and that about 7% of customers had tried 1.5x or faster. They are not a figure for any other app.

Is listening to audiobooks at 2x speed a good idea?

In Murphy and colleagues' lecture-video experiment, the cost at 2.0x was small, and Chen and colleagues' audio-only science clips showed no statistically significant effect of speed. Those were lectures and science clips, not novels, so try one chapter at 2x before you commit to a whole book.

What is the best speed to listen to audiobooks?

Start at 1.0x and step up until it stops feeling easy to follow. The studies above are the only comprehension evidence cited here, and none of them tested a full-length audiobook. Soo's view, that the best speed is the one the book was recorded at, is also a reasonable place to stay.

How long does a 10-hour audiobook take at 1.5x?

6 hours 40 minutes. At 1.25x it takes 8 hours, and at 2.0x it takes 5 hours.

Work out your own book

Take the runtime of the book you have queued, divide it by the speed you want to try, and compare that with the list above. For a worked example on a very long title, see the 11/22/63 audiobook length page.