NWNX Speech plugin!

[JeroenB]

[Blacksting]
Did you update the script, like I described in the readme.html file? If you havent it will not work, as then stilll the database is used by the nwn part, which is never filled by the plugin.

[DGunter]
No, you can only use the database option in the linux version. When the windows version becomes stable, I will port that new feature to the linux version too.

[JeroenB]

I have uploaded a new version (0.1.6). Only change in this version is that it doesn't require the libmysql.dll file anymore.

[Blacksting]

I have updated the NWN script. With a little testing the call to NWNX!SPEECH!AVAILABLE never comes back "1". Have you had a chance to test out the plugin yourself? Does it work for you?


edit:: I remembered in the back of my brain about naming the module the same as the module tag. That was the clincher. All is well

[Lanthar D'Alton]

The future is coming:
Talus 3.0 preview here:

http://lanthar.dyndsl.com/i/speechdialog.jpg

Also if you can't wait to see the new functionality available, you can try out the new /o emote method (like /dm for dm, /s does shout, /talk does talk, and /o does a self-tell) and be blown away by the ease of speaking languages and through familiars on my server at 65.68.156.73:5121

-Lanthar

[JeroenB]

I make an update to the plugin:

- No more database support, only via an internal message queue.
- Ported the system to use the STL queue class for improved stability and less dynamic memory allocation (at least at my side ).

I will upload these changes asap, and port them to the linux plugin too.

[Blacksting]

Lanthar: Does your tease line about speaking languages mean that Talus now "intercepts" traffic and can change it rather than just sniffing it? That would be damn cool.

[Lanthar D'Alton]

1: no, altering the network traffic is rather troublesome... however it handles a speech method that no one else can see.

2: Jeroen, really, just sit back and wait... If you keep giving me excuses to procrastinate, I won't get it done...

This forthcoming version alters the number of parameters for an additional value used by the new emote method. It also largely reworks the listener cpp, so any changes you make there will need to be reworked. There are several speech commands that I've added that will be in the erf. I'm also planning to add a built in command lookup std::map that may offer dialog editing on the fly... and saves to an ini file. I think I'll be leaving the permissions table in the database though, since it is player info driven.

-Lanthar

[Lanthar D'Alton]

Is that it prevents the development of an inter-server message system. I meant to develop one of these for those needing it once. Even so, people don't have to use the dll, so while I am going to finish my updated dll with the speech monitor window, I'm also going to include the standalone db interfacing exe with it, so that people can use that if they want. Similarly, I already have laid out two separate include files in the test module that I will make the erf from.

-Lanthar

[Lanthar D'Alton]

Just a note to tell all of you I've just completed the code to add the action list to the dll as a std::map<std::string, std::string> (yes, string to string, no sense keeping them as ints when nwnx only accepts strings:) ).
I already had the speech kept in a std::queue, so that's all nice and clean too. Oh, and I added the Reload Verbs button, so it's very easy to edit the comma delineated file (a generic one is generated on first run) and then hit the reload button to refresh the std::map from the file. This allows easy editing and addition of commands.

Anyway, it's quite fast, and almost ready for releasing for beta. I just need to do some final cleanup and then I'll be ready for beta...

-Lanthar

[Grinning Fool]

Slightly OT, but this seems like the best thread to use for htis. I'll be working on getting the linux version out to coincide with this (erm, the other linux version?). The linux version won't have the on-screen dialog with conversations, but will be the same in all other respects.

[Lanthar D'Alton]

I've just got to do some work on the readme file, then try it out, and then package it. The ERF has been made for the DLL plugin, and should be fully interchangeable with the db based standalone exe. I actually made two different scripts which can be swapped for each other and these specify the interfacing functions.

I need someone to help me test it... just send me an IM by icq, aim, or msn...

-Lanthar

[Lanthar D'Alton]

Still waiting for someone to message me so I can have them do a simple beta test...

I want to test while on a messenger with whoever tests it, to ensure my instructions are clear and the merging is easy...

-Lanthar

[Mr-XXS]

Is there some packets that can be missed ? Or some canals that are not detected ?

When i read my log file, i have sometimes the feeling that some sentences are missing. Is it possibile that a few sentences are not catched ?

[Lanthar D'Alton]

speech data switches to a multi-packet format when they exceed a certain size. I was able to locate the first packet, but the second seems to be in TCP/IP format instead of UDP, and I was never able to locate it... I think it *might* be sent in a compressed form.

This might eventually be addressed, but I'd rather leave the number unpublished.

-Lanthar

[Lanthar D'Alton]

No really... I need some testers! Anyone, get in touch via my instant messaging methods or by email.

ICQ: 7349572
MSN: lanthardalton@icqmail.com (email too)
AIM: lanthardalt0n@icqmail.com (note the zero)

-Lanthar