GNOME Bugzilla – Bug 455308
Output module selection with Speech Dispatcher backend
Last modified: 2008-07-22 19:28:19 UTC
The Speech Dispatcher backend currently doesn't support output module (syntyhesizer) selection from Orca preferences dialog. The module can be set in Speech Dispatcher configuration, but many users would probably like to see the available modules and be able to select them directly from orca preferences.
Created attachment 91521 [details] [review] Add support for output module setting and automatic reconnection. Speech Dispatcher 0.6.3 is needed to make the output module listing/selection work, but the backend still works with older versions without a change. Speech Dispatcher 0.6.3 will be released in a few days. CVS version may be used until then. The newer SD version allows one more feature, which is also implemented by this patch. Orca will automatically try to reconnect when the Speech Dispatcher connection is lost. This is extremely useful for example when Speech Dispatcher crashes or is just restarted. This caused Orca to hang before, but with this patch, Orca will recover automatically as soon as Speech Dispatcher becomes available again. I am including these two changes in one patch just to make it simpler, but it is not a problem to send it separately. Just let me know what you prefer. Except for these two new features, there are also a few cleanup changes. More details may be found in the Changelog entry included with the patch.
Created attachment 91549 [details] [review] Add support for output module setting and automatic reconnection A slightly improved version of the previous patch.
Created attachment 92048 [details] [review] Add support for output module and voice setting and automatic reconnection This version adds support for listing the available voices for each output module and their selection.
This patch introduces two new translatable strings: * "Default Synthesizer" * "%s default voice" If that is a real problem to accept these two strings, the first one can be replaced just by "Default", which is already present elsewhere in orca. The second string can be replaced by "Speech Dispatcher configured default voice", which was present in the previous version of the backend. The new strings make more sense in the new context, but I prefer having the functionality rolled in, rather than blocked just because of strings...
Created attachment 92194 [details] [review] Add support for output module and voice setting and automatic reconnection Futher improved version.
I don't have speech dispatcher installed/configured at the moment, so I cannot test this. But, I do notice a couple new strings being marked for translation: + _SERVER_NAMES = {DEFAULT_SERVER_ID: _("Default Synthesizer")} + self._default_voice_name = _("%s default voice") % id Can you please add notes for translators for these strings (Look for strings beginning with "# Translators:" elsewhere in the Orca codebase) and also prepare a message to go to gnome-i18n at gnome dot org that identifies these strings?
(In reply to comment #6) > Can you please add notes for translators for these strings (Look for strings > beginning with "# Translators:" elsewhere in the Orca codebase) and also > prepare a message to go to gnome-i18n at gnome dot org that identifies these > strings? Hello, I'm currently on a vacation so can't send a new patch. Could you, please, be so kind and add the following comments manually: _SERVER_NAMES = {DEFAULT_SERVER_ID: _("Default Synthesizer")} # Translators: "Default Synthesizer" will appear in the list of available # speech engines as a special item. It refers to the default engine # configured within the speech subsystem. Except for this item, the user # will have a chance to select a particular speech engine by its real # name, such as Festival, IBMTTS, etc. self._default_voice_name = _("%s default voice") % id # Translators: This string will appear in the list of available # voices for the current speech engine. %s will be replaced by the # name of the current speech engine, such as "Festival default voice" # or "IBMTTS default voice". It refers to the default voice # configured for given speech engine within the speech subsystem. Thank you very much for understanding.
I just tested this patch with the CVS version of speech-dispatcher: * Restarting speech-dispatcher while orca is using it: orca reconnects automatically * Selecting output module: tested espeak, flite and ibmtts. * Selecting voice/language with the espeak module It would be nice if this patch could be included because it improves speech-dispatcher support in orca significantly.
Created attachment 93193 [details] [review] Add support for output module and voice setting and automatic reconnection Added translation comments.
Thanks! Committed to gnome-2-20 branch and trunk. Will go out with Orca 2.19.90 today.