After an evaluation, GNOME has moved from Bugzilla to GitLab. Learn more about GitLab.
No new issues can be reported in GNOME Bugzilla anymore.
To report an issue in a GNOME project, go to GNOME GitLab.
Do not go to GNOME Gitlab for: Bluefish, Doxygen, GnuCash, GStreamer, java-gnome, LDTP, NetworkManager, Tomboy.
Bug 364529 - Allow voice styles (e.g., link voice) to be undefined
Allow voice styles (e.g., link voice) to be undefined
Status: RESOLVED FIXED
Product: orca
Classification: Applications
Component: speech
2.17.x
Other All
: Normal normal
: ---
Assigned To: Rich Burridge
Orca Maintainers
Depends on:
Blocks:
 
 
Reported: 2006-10-23 19:48 UTC by Willie Walker
Modified: 2007-01-02 21:25 UTC
See Also:
GNOME target: ---
GNOME version: ---


Attachments
Patch to fix the problem. (1.24 KB, patch)
2006-11-14 22:18 UTC, Rich Burridge
none Details | Review
Patch to fix the initial bogus pitch value. (756 bytes, patch)
2006-11-15 21:16 UTC, Rich Burridge
none Details | Review

Description Willie Walker 2006-10-23 19:48:46 UTC
Some users do not want some voice styles, such as the upper case and link voices.  As a means to help do this, we should consider allowing these voices styles to not even be defined at all in settings and then behave appropriately (i.e., fallback to the default voice) in the various places where they are used.
Comment 1 Mike Pedersen 2006-10-24 18:36:44 UTC
This would be very desirable as we would like to have multiple ways of defining how a link is presented in speech.  
Comment 2 Willie Walker 2006-11-04 14:54:33 UTC
In playing with this today, the way to make this happen with the current orca implementation is to set the voice to None in user-settings.py.  For example, change the following line in user-settings.py from:

'hyperlink' : orca.acss.ACSS({'average-pitch': 2}),

to:

'hyperlink' : None,

This will tell Orca to use the default voice.
Comment 3 Willie Walker 2006-11-04 14:58:37 UTC
Note, however, that we can't close this bug so easily.  We may want to have some better smarts to combine utterances that use the same voice style so that the prosody works out well.  This will take a bit of thought since we often want longer pauses to better delineate things.  

Furthermore, we may also want to refactor the speechgenerator to return [(utterance, acss), (utterance, acss), ...] instead of [utterance, utterance, ...] to allow it to provide finer granularity over the voice style used.  For people that don't want different voice styles, using the "None" solution described in the previous comment would continue to work.
Comment 4 Rich Burridge 2006-11-14 22:18:37 UTC
Created attachment 76599 [details] [review]
Patch to fix the problem.

By default, the hyperlink voice will now initially get it's settings
from the default voice.
Comment 5 Rich Burridge 2006-11-14 22:21:56 UTC
Changes checked into CVS HEAD. Mike could you let me know if this
is what you'd like before I close it out? Thanks.
Comment 6 Mike Pedersen 2006-11-15 16:58:36 UTC
This change seems to work as expected.  However,  when I removed my .orca directory to test this, my pitch defaulted to 9.  As I've never seen this before I didn't know if it was some how an unintended change.  
Comment 7 Rich Burridge 2006-11-15 18:19:48 UTC
"my pitch defaulted to 9"

Do you mean pitch or gain? The three default settings start with:

    defaultACSS[acss.ACSS.RATE] = 50
    defaultACSS[acss.ACSS.GAIN] = 9
    defaultACSS[acss.ACSS.AVERAGE_PITCH] = 5
Comment 8 Mike Pedersen 2006-11-15 20:39:21 UTC
I mean pitch.  To test:  Delete your .orca directory, start orca, enter the config UI.  When you tab to the pitch at least if you are using Dectalk it seems to be set to 9.  
Comment 9 Rich Burridge 2006-11-15 21:16:56 UTC
Created attachment 76679 [details] [review]
Patch to fix the initial bogus pitch value.

Twas in the Glade file. Change checked into CVS HEAD.