2024-05-27, 02:47
Hi. Would it be possible to ad NVDA support to windows? There is a version of the screen reader addon already supporting this. I can't wait to try this one out though. Thank you for working on this!
2024-11-08, 10:22
[font][font]Hello! Is this continuing to develop?[/font][/font]
2024-11-09, 13:25
I am still working on this. I know that many people are waiting. I am putting lots of time into it and hope to have a Beta this year.
2024-12-26, 16:48
Hello,
I am interested in the project and would be happy to help. I can't get the code in the repository to run on Kodi v21.1 (Omega) on Windows 10. Would you be able to provide a working Kodi version? Then I could work on it and continuously update the code on GitHub.
best regards
I am interested in the project and would be happy to help. I can't get the code in the repository to run on Kodi v21.1 (Omega) on Windows 10. Would you be able to provide a working Kodi version? Then I could work on it and continuously update the code on GitHub.
best regards
2025-01-12, 17:40
Glad for the offer. We should figure out how we can best work together. At the moment I'm working hard to get at least an alpha out.
Current status:
Linux support of googleTTS and eSpeak engines. Also, mpv, mplayer, and sfx (kodi built-in) players.
Windows is my secondary platform (and most people's primary) supports eSpeak, mpv (but not in slave mode yet), mplayer.
Caching, custom voiced configuration, detection and repair of invalid settings are in fair shape on both platforms.
Designed to be multi-lingual, but limited testing.
Currently working hard to get Windows in good shape. Having to improve handling of unavailable players. Should take several days.
Succesfully prototyped using trivial powershell script from python to use the Windows 11 TTS. Will add as an engine/player after I resolve unavailble player handling.
Once this is done the focus will be testing and fixing bugs. Also focus on adding important content to the baby configuration help system.
After release much cleanup/refactoring needs to be done. I exploaded the size of this project. Learning Python as well as technical apsects about the project contributed to the mess.
Enough of that. What I could use help on include:
Windows experience (I am no expert). In particular, .NET and Powershell
Windows TTS experience. Knowledge of popular TTS engines and services would be great. I haven't touched SpeechDispatcher nor JAWS. There are probably some services that people pay for to get better voices and features.
Getting mpv player to work in slave mode (uses named sockets, but probably can do with regular sockets). I had it working with a hand-built version of Kodi using a newer version of Python (probably python 3.10) but the standard Kodi uses Python 3.6 (I think). Anyway, this python does not have support for mkfifo. On Windows Kodi is built with Python and because they renamed several libs you can't just use any old library.
Let me know if you are aware of a small, good, efficient mp3 player, preferably built into base Windows 11 AND supports a type of slave mode where you can keep the player alive and through some other socket/file issue commands to play another file, to stop playing current and to change playback speed, volume, etc.. MPV does this (well, it is not so small, and the slave mode is not perfect, but does the job). I find using a cache and slave mode very helpful because it greatly improves the responsiveness.
Knowledge of Kodi's scripting (specifically UI controls) I find the Kodi controls a bit difficult to work with. There are lots of warts and suprises.
Knowldege of good general UI practices for accessibility would be great. I know some, but it is mostly based on Java Swing from 20 years ago. I'm slowly creating a means to add meta-data to existing Kodi .xml files to give hints to my .xml parser. Eventually I would like to use this to help get Kodi's scripting language altered to be more accessible.
That is about all I can think of. There is plenty of work to do. The biggest impediment to sharing the work is that I need to stablize things and and be more disciplined in making changes (I tend to make massive changes). At least there is decent separation between the code that does 'screen scraping', the code that caches, the code that handles phrases, from the rest of the code. After a good clean-up it will be much better.
I am open to suggestions and questions. I'm just a bit slow to respond.
Current status:
Linux support of googleTTS and eSpeak engines. Also, mpv, mplayer, and sfx (kodi built-in) players.
Windows is my secondary platform (and most people's primary) supports eSpeak, mpv (but not in slave mode yet), mplayer.
Caching, custom voiced configuration, detection and repair of invalid settings are in fair shape on both platforms.
Designed to be multi-lingual, but limited testing.
Currently working hard to get Windows in good shape. Having to improve handling of unavailable players. Should take several days.
Succesfully prototyped using trivial powershell script from python to use the Windows 11 TTS. Will add as an engine/player after I resolve unavailble player handling.
Once this is done the focus will be testing and fixing bugs. Also focus on adding important content to the baby configuration help system.
After release much cleanup/refactoring needs to be done. I exploaded the size of this project. Learning Python as well as technical apsects about the project contributed to the mess.
Enough of that. What I could use help on include:
Windows experience (I am no expert). In particular, .NET and Powershell
Windows TTS experience. Knowledge of popular TTS engines and services would be great. I haven't touched SpeechDispatcher nor JAWS. There are probably some services that people pay for to get better voices and features.
Getting mpv player to work in slave mode (uses named sockets, but probably can do with regular sockets). I had it working with a hand-built version of Kodi using a newer version of Python (probably python 3.10) but the standard Kodi uses Python 3.6 (I think). Anyway, this python does not have support for mkfifo. On Windows Kodi is built with Python and because they renamed several libs you can't just use any old library.
Let me know if you are aware of a small, good, efficient mp3 player, preferably built into base Windows 11 AND supports a type of slave mode where you can keep the player alive and through some other socket/file issue commands to play another file, to stop playing current and to change playback speed, volume, etc.. MPV does this (well, it is not so small, and the slave mode is not perfect, but does the job). I find using a cache and slave mode very helpful because it greatly improves the responsiveness.
Knowledge of Kodi's scripting (specifically UI controls) I find the Kodi controls a bit difficult to work with. There are lots of warts and suprises.
Knowldege of good general UI practices for accessibility would be great. I know some, but it is mostly based on Java Swing from 20 years ago. I'm slowly creating a means to add meta-data to existing Kodi .xml files to give hints to my .xml parser. Eventually I would like to use this to help get Kodi's scripting language altered to be more accessible.
That is about all I can think of. There is plenty of work to do. The biggest impediment to sharing the work is that I need to stablize things and and be more disciplined in making changes (I tend to make massive changes). At least there is decent separation between the code that does 'screen scraping', the code that caches, the code that handles phrases, from the rest of the code. After a good clean-up it will be much better.
I am open to suggestions and questions. I'm just a bit slow to respond.
