Desi Mazdur speaks about searching web pages with speech recognition software

Desi Mazdur talks to us about having RSI and relying on speech recognition software to navigate the web.


Nic: Hi. I’m Nic Steenhout. You’re listening to the Accessibility Rules Soundbite, a series of short podcasts where people with disabilities explain their impairments and what barriers they encounter on the web.

This week, I’m speaking with Desi Mazdur. Desi has a background as a programmer, but has not been able to work for a little bit due to his disability. He’s an avid user of speech recognition and he’s going to tell us a little bit about that in a bit.

Hi, Desi. How are you?

Desi: I’m good. How are you?

Nic: I’m doing pretty good. You’re telling me that you use speech recognition. What’s your disability? What’s your impairment?

Desi: I’m suffering from pain in the fingers and that radiates. It goes all the way to the neck, so it’s difficult for me to type on the keyboard or use a mouse. Depending on the doctor I visit, it’s either called repetitive stress injury problems or also fibromyalgia. Those are my difficulties in using the computer. In addition, I’m also not able to sit or stand for more than 10 minutes at a time, but that’s kind of different. It creates difficulties, but not that much, which is the typing on the keyboard or using a mouse.

Nic: Fair enough. Tell us a little bit about the barriers you encounter on the web when you’re trying to do day-to-day activities on the web. What kind of problems do you find?

Desi: There are actually plenty of problems. For example, let’s say Facebook is … obviously, everyone knows Facebook, but sites like Facebook, Pinterest, social media, they assume that a user is going to have access to a keyboard and he or she is going to press on certain keys and only then the Facebook code will execute. For example, let’s say I want to search with some keywords. Let’s say “Donald Trump” in Facebook. Now, what Facebook is looking for is it is looking for the user to press a key, “D,” the first letter “D” and then the letter “O” and so on. The reason, I believe, is that Facebook is … as soon as the user presses the letter “D,” Facebook starts searching for all the results that are beginning with the letter “D.”

Then you press the letter “O,” then Facebook starts searching for all the results with the letter “DO,” but in the case of speech recognition program, there is no such keyboard events. The speech recognition program which I am using, which is Dragon NaturallySpeaking, it is going to just put in all these letters, the spelling of Donald Trump, it is going to put all of it at one go in the Facebook search box. From the Facebook perspective, nothing has been entered into the search box, so the Facebook code has not seen any letters being entered, while from the user’s perspective, I have already entered the word “Donald Trump.” Now, when I press on the “Enter” key to search, Facebook doesn’t search anything because it still says there is no content.

Nic: That’s a big problem. Do you encounter that in a lot of places, or it is truly just a handful of websites that have that problem?

Desi: I think it isn’t all places. I’ve been a programmer, but I have not been a web programmer, but I think all the places where there is an assumption or there is a functionality that we should start searching for results as soon as the user starts pressing a key, people suffer from this. Pinterest is also searching from this issue. Google should also suffer, but what I’ve seen is Google somehow recognizes. Let’s say if I enter, to use the earlier example, if I enter the word “Donald Trump” and press “search,” Google does search for the words “Donald Trump.” I am not sure why Facebook and Pinterest don’t.

Nic: That’s an interesting perspective that I actually had not heard about, so I’ll be looking into that. Thank you for that. Desi, if there was one thing you would like to get designers or developers to remember about accessibility, what would that be?

Desi: Actually, since I have been a developer earlier, I think well, obviously, the web designers should be aware that if they put accessibility right from the starting of their project, that is going to be much better, but what I have seen is that maybe with efforts like yours, we are at a stage in 2018 where there is a good amount of awareness with the web designers, but the tools that they’re using, those tools are not, for lack of a better word, accessibility, accessibility feature compatible.

It’s more … instead of the web designers, I would say that the people who are creating tools, they need to be aware that downstream when their end users were web designers, when they are going to design something for their end users, which is website owners and then eventually people like me, the end consumer, we are not going to find this useful and again, when we go up the chain, that ultimately leads to a loss in revenue for the website owner as well as the web designer and then the tools developed. I guess that’s where the problem may be.

Nic: Well, thank you for that short chat just to give a quick soundbite about accessibility and give people a new perspective on some issues with using speech recognition to interact with computers. Desi, I really appreciated this chat. Thank you and hopefully, you have a good rest of your day.

Desi: Thanks a lot.

Nic: That’s it for now. Thanks for listening.

If you enjoyed this Accessibility Soundbite, please support the show at