Scott Small π¨π¦Question for people using screen readers:
Assuming proper camel-casing was used, can your screen reader pick up on the difference between the #SuperBowl and #SuperbOwl hashtags?
Yes, they sound different
No, they sound the same
Poll ended at .
R. L. Dane 
@RL_Dane I've seen a few toots about owls today and I'm rather curious! Especially between different versions of screen-reader software - I'm sure _some_ will get it right, but I'm wondering just how universal that is.
Just FYI, flite (festvox lite, a very basic FOSS speech synthesis program) pronounces SuperBowl and SuperbOwl the same (as super bowl). I don't think it understands camelcase at all, though.
ROFL, yeah, it doesn't. It just pronounced ExpertsExchange the way you might think!!
@RL_Dane LOL - ExpertsExchange... haven't thought about that place in years πβ
andwhyisit
@RL_Dane @smallsco For anyone not familiar with the Coen pub: https://www.news.com.au/national/sexchange-hotel-in-coen-queensland-a-tourist-hotspot/news-story/a8e5b565b60d9bacd2ec0e3268c978b4
@andwhyisit @RL_Dane Thanks for the context! This is a great story and made me laugh πβ
BadExampleMan π£
Lara J. Martin@RL_Dane ooo do you actually use Flite? I haven't heard of many people using it. I'm voice LJM!
@laramar
No way! That's too cool π
I've used it a little, although I can't find a list of phones/diphones/phonemes to use with it anywhere.
I had been using mimic, which is a *little* more natural-sounduing, but I couldn't find it in #OpenBSD ports at that moment, and since that was the box I was on right then, I went with the next best thing.
Flite doesn't sound too bad, it's just way too quiet. How do you select a voice? There's not much in the way of online help.
@RL_Dane
Yeah! I used to be a masters student for Alan Black (voice awb), and I persuaded him to let me record myself. it didn't take much because apparently not too many people want to sit in an airless soundproof both for several hours.
It's been a while (6 years?) since I used Flite on my phone, but I know you can download the voice files separately.
Oh, that's neat! Yeah, I understand you have to read a LOT of random text to get a decent sample set of all possible phonemes.
I used to love playing with Macintalk on the classic Macs. Always thought it had a slight Indian accent ;)
Cowgirlcoder@cowgirlcoder @RL_Dane Will do! In the mean time please boost it for visibility πβ Right now there's just a single vote, which is for the Yes option.
Leianora@likelyjanlukas @smallsco Yes. The words are pronounced different because they are... they are spelled differently, are they not?
@EJGilbert @likelyjanlukas No, the spelling is identical! The only difference is that in #SuperBowl the S and B are capitalized, but in #SuperbOwl the S and O are capitalized.
@smallsco @likelyjanlukas Cool! I could have gone through and read that letter by letter and the pitch of the spoken letter would tell me if the case was upper or lower, but I am lazy. :D
@stephpilon086 @smallsco Absolutely. I haven't noticed any issues with super bowl vs. superb owl. Definitely appreciating the humor.
Cassana@smallsco Definitely a difference here. And, SuperbOwl FTW! What really isn't funny is how stubborn autocorrect is about changing the spelling back to the other thing, making me want to erase that entry in the dictionary, or disinherit my phone. Anyway, another fun hashtag like that was #MusicalBumParty, umm, #MusicAlbumParty.
@smallsco Yes, tested with NVDA and five different speech synthesizers.
Monique@smallsco Yes, it sure can. In the second example, the pronounciation is all off, Instead of the bowl, there's suddenly an owl in there... lol.
Matthew Cooper@smallsco yes. Can see the difference in the hashtags with a screen reader. Using VoiceOver on iOS. When you capitalise each word, generally it tries to treat it as a different word so just imagine a space between the last lowercase and next uppercase letter. Apologies for demonstration on how to suck eggs if applicable.
Macer@smallsco People have too short an attention span to piece that together in their brains.
Willem Van den Ende - Writing@smallsco (minus one for this mobile phone user who voted by accident for "same")
Haelwenn /ΡΠ»Π²ΡΠ½/ 
@smallsco Sound different but the prononciation is wrong :/
Andrew Woods@smallsco Looks like it does make a difference for screen readers. Additionally, camel casing hashtags provides clarity for sighted users too - as noted in the replies with humorous examples
@awoodsnet We're at 70% Yes, 30% No right now with 43 votes, so it definitely makes a difference if you're using the right software πβ
Enaraπ©βπ¦―π@smallsco Mine can. Just to let it know, I use NVDA and Windows 11 on computer. And VoiceOver on iOS does it as well.
Denez Van Dyck βTested on macOS:
- Using an American voice, both hashtags are pronounced "superbowl".
- Using a British voice, I can clearly hear the difference between "superbowl" and "superb owl".
@dvd That's fascinating, I didn't even think about the possibility of changing the voice, just the software.
I wonder how many of the "No" votes are using an American voice?
Kagan MacTane (he/him)@smallsco Yes, they sound different, using NVDA on Windows 10 with Firefox.
Christoph Thiede 
@smallsco If you really to want other people (who don't use screen readers) to learn something from this questionnaire, please include a neutral option to the questionnaire so we can also see the results.
@LinqLover I didn't want a "See the results" option to skew the results. You can bookmark the poll and return when it closes (12 hours from now) to see them!
I'm also happy to give an update if asked - we're currently at 70% Yes, 30% No, with 66 people voting.
@smallsco
I don't normally use a screen reader, but I had to try it out, with these results from the built-in Google screen reader: with English or French reading language selected, they sound the same. With Japanese selected, they sound different! Surprisingly, both French and Japanese language versions of the paragraph were perfectly understandable, though highly accented.
I don't normally use a screen reader, but I had to try it out, with these results from the built-in Google screen reader: with English or French reading language selected, they sound the same. With Japanese selected, they sound different! Surprisingly, both French and Japanese language versions of the paragraph were perfectly understandable, though highly accented.
May Keable π³οΈβππ³οΈββ§οΈ@smallsco I don't use one normally but for funsies I tried the built-in Android Google Assistant one. She could not tell the difference.
Laura Langdon@smallsco I use Natural Reader for narration of long-form text (not social media) for #Neurodivergent reasons, not for visual accessibility, and when I pasted the text into it there was no audible description between the two hashtags.
lipfang π³οΈββ§οΈ@smallsco Just tried it with Android's TalkBack and they sound the same!
@smallsco @munchkinbear my if Camel casing is initial caps and words smushed together, then yes it's discernable.
@ladymunch @munchkinbear Yeah, it's when you capitalize the first letter of each new word instead of using spaces to delimit words. I'm not actually sure why it's called "camel case" though!
@smallsco it's weird for sure. I just call it Initial Letter Casing but that's me.
Amanda Carson@smallsco Yes. It picks up the difference demonstrated in this post. Hope this is helpful.
@smallsco Mine can, yes.
Ahimsa@smallsco I just did a test with the built-in screen reader on Mac (VoiceOver) and they sounded different.
Nick Giannak III
Zy 
@smallsco
TalkBack sur android change pas la prononciation
TalkBack sur android change pas la prononciation
tarper24Using Talk Back on Android, it sounds like super bowl no matter the casing, much to my disappointment
Melvin Gundlach@smallsco VoiceOver on the iPhone differentiates.
Jan Thie@Jantar Mastodon doesn't have a way to see how the vote is developing if you haven't voted yourself - but feel free to ask me for an update any time, as long as I'm not in a work meeting I'll post one.
We're currently at 71% Yes, 29% No, with 139 people voting.
@smallsco
Thank you for those scores.
That will do me for now - and with such a spread it's hardly to be expected things will change much, I'd think?
Anyway, I will now just wait till the end of the poll, but thank you very much for the kind offer!
Thank you for those scores.
That will do me for now - and with such a spread it's hardly to be expected things will change much, I'd think?
Anyway, I will now just wait till the end of the poll, but thank you very much for the kind offer!
Kit@smallsco I'll add that my screenreader tries so hard to be helpful and makes many delightful gaffes which I've come to enjoy:
"No." --> "number"
so a passage in an essay:
"Is that what makes it a good story? No. Is that what makes it a well-told, well-structured story? Fuck no. But many assume it does."
becomes:
"Is that what makes it a good story? Number! Is that what makes it a well-told, well-structured story? Fuck number!"
π€£π please be sure to engineer more of this joy for me
e. hashman 
@smallsco just tested it, no difference. However, it definitely pronounces "super bowl" and "superb owl" (with spaces) differently. Using Android Select to Speak
mhoye@smallsco doling out a Fry Squint for that βboost for visibilityβ line over here
@mhoye LOL, I didn't even think of that πβ
I mean, people can use screen readers for other reasons, too πβ
Dan Keck@smallsco
Sounds the same on my phone: TalkBack on Android. Both times I hear "Super Bowl." I tried a few different voices.
Sounds the same on my phone: TalkBack on Android. Both times I hear "Super Bowl." I tried a few different voices.
@smallsco very curious about the results... And I don't know how mastodon works yet, so maybe replying with man I get notified when it's done?
@mattgood If you voted on it, you'll automatically get a notification when the poll closes (in just over 2 hours from now).
Alternatively, if you didn't vote but are just curious about the results, you can bookmark the post and come back to it when the poll closes, it will update to show you the results.
We're currently at 72% Yes, 28% No, with 178 people voting.