Re: [orca-list] How to write extension for ibus input method?

[Cameron Wong <hgneng gmail com>]

Hi Joanie,

I made a Linux distribution (eGuideDog Linux) for Chinese blind some years ago: http://www.eguidedog.net/linux.php

It's the only accessible Linux for Chinese blind. The Linux distribution is based on Ubuntu LTS. I install Chinese TTS in it, hack input method to be readable and make some essential configuration for Chinese locale in the distribution. I also wrote some tutorials to teach blind people how to install and use it.

I am the main contributor of Chinese support of eSpeak and author of Ekho TTS (http://www.eguidedog.net/ekho.php). These are the only two open source TTS that support Chinese.

The description code were not in speech-dispatcher or any TTS. It locates in ibus. I hack ibus and re-install it in my Linux distribution. The description is not standard and to be improved. But I don't think there a standard.

I would like to repleased a new version of eGuideDog Linux at April after Ubuntu 18.04 released. Then I will improve the description. It needs much of work because I need to go though all Chinese characters. When the description is good enough, I think it would be better to push it to Orca so it become default part of Orca.

Cameron

2018-03-27 23:55 GMT+08:00 Joanmarie Diggs <jdiggs igalia com>:

Hi Cameron.

I'll keep trying to reproduce the crash. Regarding this:

> I suddenly remember that I haven't reply question above before I go to
> bed. I hacked ibus. In the refreshing event handler of ibus panel, I
> called spd_say to speak description of the highlighted character. At
> that time, Orca speak nothing about ibus panel, so it works. But now I
> can't make ibus and Orca speak at the same time.

So to be sure I understand what you're saying, the descriptions are not
anywhere in speech dispatcher's code or in the code of the synthesizer
you are using. Instead, they are in a local file that only you have. Is
that correct?

And are your descriptions something standard that all blind,
speech-output-using readers of Chinese characters would recognize? Or
are they something you came up with, but are not standardized anywhere?

I'm interested in knowing these things because I'd like to see
accessible Chinese input available for all Chinese Linux users who are
blind. How much of that belongs in speech-dispatcher versus Orca, I
don't know. But if we can figure it all out, wouldn't that be cool? :)

Thanks again for all your help and patience on getting this right!
--joanie