Hi list, A quick morning post. I'll supply you with the exact prompts later today if it is needed: I'm trying to set up both brlTTY and the gnome-orca package and am running into a similar problem. I've written the BrlTTY config for my Voyager display. When running brlTTY, however, it brailles that it cannot find a text screen at all. I'm using a serial console connected to a Windows terminal emulator on the same machine. In that emu, WIndows speech and Braille work well enough. HOwever, I'd still like to use brlTTY in GNome apps like Orca, Gnopernicus or the terminal without having to give up my serial console. Any tips as to what i need to do?
Orca seems to face a similar problem when I run it via the serial console. It blurts out a stack backtrace essentially saying that it cannot find the screen. I've also got other issues with Orca: As Festival still doesn't work, I've edited my speech dispatcher config to use eSpeak by default for English. UPon running orca in the Gnome terminal it prints and speaks that it could find the Linux accessibility system and affter that it comes up with no on-screen or spoken prompt. It just sits there waiting for input and I can type in many lines of it from STDIN without anything happening on screen. Hmm, is Orca designed to work with eSPeak initially as opposed to Festival? If all else fails, is there some Python script or config I can hack at to do all the initial setup things manually. If I run Orca directly in the Gnome run box, it starts but I can see no on-screen manifestation of the app nor nothing in the alt+tab dialog. Pressing ins plus space appears to do nothing, I can see no visible change nor hear any dialog appearing. Other than that Orca seems to track the focus and prompts much more sensibly than Gnopernicus does. Which brings me to my last point, is there some file in which I can change the hotkeys? Better yet would be a laptop specific layout. It is no fun trying to press fn, ins and some piece of the emulated laptop numpad on an HP machine at the same time. Many WINdows readers offer laptop layouts. Gnopernicus has trouble with speech, too. I've written a new speech dispatcher config file which should utilize the full pitch, speed, volume and voice range. Yet only speed and volume parameters in the absolute edit mode affect matters. Toggling the voice or changing the annoyingly high pitch does nothing. I have not looked into the logs yet. But might this be a Gnopernicus issue? I'm more inclined to think it is something in my speech Dispatcher config file at the moment. Oh well, a GUI front end and ssmart apps would prevent such issues. Alan Cooper's views on considerate and smart software come to mind. Last and not least a quick Q about the Gnome file manager Nautilus: Are there accessible alternatives to it in the spirit of Xplorer 2 or Total COmmander? It uses the icon view by default whose hotkeys require that you can see how icons are layed out on screenTHis is less than ideal for me. There's also a details like view but unlike in WIndows, one isn't able to edit the order and visibility of individual columns, oh well. This is a must if you'd like to get decent output with Gnopernicus. Gnopernicus also prefixes every terminal prompt and list item with some initial string, that slows down traversing them. In addition to a good details view I would need a plain list view with only the file name in it. It would be great if you could kill the icons in that, too. Icons add very littel in my GUI experience unless there are large blobs of difference in hue, saturation or luminance. THe actual shape of the icon doesn't matter that much, in a way. Here's my latest speech dispatcher config for eSPeak: espeak-generic.conf: # Supports pitch, speed and volume. GenericExecuteSynth \ "echo \"$DATA\" | speak -v $VOICE -s $RATE -a $VOLUME -p $PITCH --stdin" # Includes all the US English. GenericLanguage "en" "english" # GenericLanguage "fi" "finnish" # The naming of some of the untitled US English voices is a bit ad hoc, oh well. AddVoice "en" "Default" "en" AddVoice "en" "Echo" "en1" AddVoice "en" "fuzzy" "en2" AddVoice "en" "English (3)" "en3" AddVoice "en" "Male (4)" "en4" AddVoice "en" "blocked" "en6" AddVoice "en" "Male (7)" "en7" AddVoice "en" "Old" "en8" AddVoice "en" "Croak" "en-croak" AddVoice "en" "Female" "en-f" AddVoice "en" "Lancashire" "en-n" AddVoice "en" "Female (North)" "en-n-f" AddVoice "en" "English rp" "en-rp" AddVoice "en" "English rp-f " "en-rp-f" AddVoice "en" "English wmids" "en-wm" AddVoice "en" "En-wm-f" "en-wm-f" # Here are the multipliers and offsets for exposing the maximum range and resolution from eSpeak. # dispatcher gives values from -100 to 100 and eSpeak supports: # volume 0 to 200, pitch 0 to 99, rate 80 to 320 Words per min # Thus we derive the fixed-point scaling factor (last two digits fractional), and the offset: GenericRateAdd 200 GenericRateMultiply 120 GenericRateForceInteger 1 GenericPitchAdd 100 GenericPitchMultiply 49 GenericPitchForceInteger 1 GenericVolumeAdd 100 GenericVolumeMultiply 100 GenericVolumeForceInteger 1 Debug 0 Which reminds me, do either Gnopernicus or Orca support switching the language of screen text say from English to Finnish? I'm trying to prototype the Finnish eSPeak support and iwould like to use it in the real world such as in e-mail or the Web. SOrry if this mail comes across as more negative than usual: Getting the speech support working was the big win to me but now I'm having a handful of minor but nevertheless annoying issues I'd lie to fix as soon as possible. None of them actually prevent me from using Linux, which is good all in all. But I gotta go now. -- With kind regards Veli-Pekka Tätilä ([EMAIL PROTECTED]) Accessibility, game music, synthesizers and programming: http://www.student.oulu.fi/~vtatila/ -- Ubuntu-accessibility mailing list [email protected] https://lists.ubuntu.com/mailman/listinfo/ubuntu-accessibility
