Providing text to speech readback of a string of numeric digits as individual digits rather than as a number would indeed be useful. However, I'm not so sure that's a Speakup problem, meaning that I suspect this is a bug with the TTS itself. Historically, TTS engines have assumed responsibility for similar content interpretations. This has always been problematic and inappropriate, imo, but getting them to understand that has proven problematic in my experience. My favorite example is the way Eloquence (and it's Linux versions) insisted on rendering cd as "cendelas," not change directory, not compact disc, not even certificate of deposit, but candelas as if we blind folks were exclusively fixated on measuring luminosity in some arbitrary direction! Rendering strings of numeric digits as digits would actually satisfy the more common use case, imo. I find it annoyingly difficult to translate a phone number from a speech rendering that starts with so many billion, followed by so many million, and then so many thousands. And why? Just because some developer saw an opportunity to show off how clever they were? But to go back to my original point, I don't believe Speakup is processing any semantic understanding of the text it feeds to the speech engine. And, that's what it would take to solve this problem in Speakup. I could be wrong, of course. Best, Janina Chime Hart writes: > Hi All: I think Speakup is an only screen-reader in almost any platform > without an option to switch to hear single digits while reading. Maybe > Chromevox may not have this, but just about all others from DOS up through > Fenrir have offered this in some form. While it may seem like a small thing, > while reading an Alpine mail index, hearing the word "hundred" feels as if > it wastes alot of time. If it would be more official I can file a wish-list > bug against speakup-tools? When I run reportbug, that seems an only package > to file against. And speaking of outstanding bugs, back on February 20 I > filed > #1062507 > about the DecTalk drivers. Funny thing was, even while reading over what I > had submitted, settings dropped. I looked around the Speakup drivers > directory in a newest 6.9 kernel, where a specific change which Samuel had > helped me with reguarding flush time. 10 is wonderful but in dectlk.c it > still says 4000. > In basicly nearly 21years of useing Speakup, an only other wish list item > would be an exception dictionary. Again, an item in most other > screen-readers. Thanks so much in advance for listing-and-considering these. > Chime -- Janina Sajka (she/her/hers) Accessibility Consultant https://linkedin.com/in/jsajka The World Wide Web Consortium (W3C), Web Accessibility Initiative (WAI) Co-Chair, Accessible Platform Architectures http://www.w3.org/wai/apa Linux Foundation Fellow https://www.linuxfoundation.org/board-of-directors-2/
