W3C home > Mailing lists > Public > w3c-wai-ig@w3.org > April to June 2017

Re: Screen Readers | Rate = Words Per Minute

From: Léonie Watson <tink@tink.uk>
Date: Wed, 10 May 2017 19:01:06 +0100
To: "Wright, Isaiah" <Isaiah.Wright@ally.com>, Chaals is Charles McCathie Nevile <chaals@yandex-team.ru>, "w3c-wai-ig@w3.org" <w3c-wai-ig@w3.org>
Message-ID: <6624f0a1-1aed-73d6-c255-376726d15242@tink.uk>
On 10/05/2017 14:25, Wright, Isaiah wrote:
> Currently I’m making my own screen recordings using NVDA but I’m having
> troubles gauging rates and what falls into low, medium, or high
> categories, especially since there isn’t a rate percentage to WPM
> conversion chart.

What is it you actually want to demonstrate? Different people will 
listen to their screen reader at different speeds, and they will change 
the speed depending on the thing they're doing at the time.

To illustrate this... when I first used a screen reader I left it on the 
default speaking rate because I found it difficult to understand. As I 
became used to listening to it, I gradually started to increase the 
speed. Now I use NVDA at 100% if I'm just skimming emails or browsing 
the web, about 80% if I'm reading something I want to pay attention to, 
and about 60% if I'm reading something I want to digest properly. If I 
reduce speed to 40% it is too slow to be bearable, and anything slower 
than that leaves me shouting in exasperation.

There are other factors: some screen reader voices are better (more 
accurate) and more performant at higher speeds; environmental factors 
like noise can make listening at higher speeds more difficult; if the 
thing I'm reading is not in English it's easier to slow down the 
speaking rate because my comprehension isn't as good.

So it's a great idea to demonstrate the way someone might listen to a 
screen reader, but I wouldn't worry about how fast it actually is or how 
many WPM it equates to, because those are generally meaningless in this 

@LeonieWatson tink.uk Carpe diem
Received on Wednesday, 10 May 2017 18:01:42 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 10 May 2017 18:01:42 UTC