Requests for additional extensions (?)

Use of external libraries (DLLs) from ZGE.

Moderator: Moderators

User avatar
Kjell
Posts: 1668
Joined: Sat Feb 23, 2008 11:15 pm

Post by Kjell » Wed Feb 11, 2015 4:33 pm

Hi Rado1,
Rado1 wrote:Hmmm... the problem is mapping of sprite positions, rotations and sizes to 3D skelet. And also a good editor. But I'm not 100% sure, maybe I'll do some more exploration of this possibility.
You can do 2D skeleton animation in any 3D software just fine. Here are a couple of examples ...

http://www.youtube.com/watch?v=yI2eh-Dv6JY ( Maya )
http://www.youtube.com/watch?v=M_ZL4zR9Urw ( 3ds Max )
http://www.youtube.com/watch?v=rczDMXBoCwo ( Blender )

K

User avatar
Rado1
Posts: 770
Joined: Wed May 05, 2010 12:16 pm

Post by Rado1 » Wed Feb 11, 2015 4:50 pm

Ok I'll have a look at the examples and maybe to extend the library to support also 2D animations (if Cal3D will allow it) - e.g. to obtain positions and rotations of bones at given time frame, their links to materials, etc.

User avatar
VilleK
Site Admin
Posts: 1946
Joined: Mon Jan 15, 2007 4:50 pm
Location: Stockholm, Sweden
Contact:

Post by VilleK » Fri Feb 13, 2015 8:55 am

This is pretty interesting: stb_truetype at https://github.com/nothings/stb

It is a truetype font parser and renderer, implemented as a single C-header file. Example use here.

It could be used to build a module for ZGE to support nice fonts cross-platform.

Some of the other libs at that site looks cool too. Very clean and small implementations.

Imerion
Posts: 197
Joined: Sun Feb 09, 2014 4:42 pm

Post by Imerion » Thu Feb 19, 2015 12:46 am

I just got an idea for an extension I might find use for : eSpeak.
http://espeak.sourceforge.net/

It's a speech synthesis software which can create voices in many different languages without using recordings. This keeps it very small and compact. Settings such as speed, pitch, etc can be changed.

I think this might be useful for voices in arcade games and such. eSpeak can generate wav-files, so an extension is not really necessary. But it would help keep program size down and perhaps provide some nice flexibility.

Just a thought. It's written in C, but I have no idea how hard it would be to turn it into a ZGE-compatible library.

User avatar
rrTea
Posts: 333
Joined: Sat Feb 15, 2014 9:54 am

Post by rrTea » Thu Feb 19, 2015 1:08 am

Imerion wrote:I just got an idea for an extension I might find use for : eSpeak.
http://espeak.sourceforge.net/
I'm also interested in this (it was discussed in the chat too). Even if it weren't as advanced as the one from the link (Amiga's Workbench 1.3 "say" level would be more than enough for me) I'd be super happy to use it!

An example of a relatively recent simple game where this was put to good use is Mitchell's Wakugumi: this video doesn't really show it but I was unable to find better videos - just grab the game if you can, it's really good and cheap anyway.

User avatar
Rado1
Posts: 770
Joined: Wed May 05, 2010 12:16 pm

Post by Rado1 » Fri Feb 20, 2015 8:24 pm

In meantime, I tried eSpeak and here's my observation: it produces very bad results - robot-like speech, hard to understand; there's a lot of criticism around. The only usable voices of their examples come from Microsoft SAPI5, so why not to use built-in Microsoft Speech directly?

Attached you will find an initial demo (created in 20 minutes), based on my own ZgeSpeak.dll - a small library making use of Microsoft Speech API. I hope you will enjoy it.

BTW there's a native TTS support also on Android and luckily it is included also in NDK, so ZgeSpeak can be extended also to Android, if you want.
Attachments
demo1.zip
Speech synthesis in ZGE - the first demo
(4.22 KiB) Downloaded 77 times

User avatar
Rado1
Posts: 770
Joined: Wed May 05, 2010 12:16 pm

Post by Rado1 » Sat Feb 21, 2015 12:59 pm

A small correction of the previous demo, because calling zsFinish() twice resulted in access violation error.
Attachments
demo1.zip
Corrected demo.
(4.22 KiB) Downloaded 80 times

Imerion
Posts: 197
Joined: Sun Feb 09, 2014 4:42 pm

Post by Imerion » Sat Feb 21, 2015 1:16 pm

eSpeak varies a lot depending on language, and it often sounds very robotic. The whisper-setting is really cool though, and could be used in horror-games, for example. :)

This sounds like a nice solution, but I guess it can't be used on Linux since it uses Microsofts speech api? If you add Android support as mentioned I might find use for it though!

User avatar
Rado1
Posts: 770
Joined: Wed May 05, 2010 12:16 pm

Post by Rado1 » Sat Feb 21, 2015 1:32 pm

Imerion wrote:This sounds like a nice solution, but I guess it can't be used on Linux since it uses Microsofts speech api? If you add Android support as mentioned I might find use for it though!
Hi Imerion, unfortunately all my extensions work only on Windows and Android, I do not use Linux. I'm going to have a look at TTS on Android.

Carver413
Posts: 9
Joined: Sun Apr 20, 2014 10:07 pm

Post by Carver413 » Sun Feb 22, 2015 2:53 pm

Imerion wrote:eSpeak varies a lot depending on language, and it often sounds very robotic. The whisper-setting is really cool though, and could be used in horror-games, for example. :)

This sounds like a nice solution, but I guess it can't be used on Linux since it uses Microsofts speech api? If you add Android support as mentioned I might find use for it though!
there is also festival, flite and open mary tts that work in linux. using shell scripts you can create sound files. not as good as an api but it works. flite is my favorite at the moment its small and there are 11 voices you can get for it. open mary has some nice sounding female voices but can be a bit buggy at times.

User avatar
VilleK
Site Admin
Posts: 1946
Joined: Mon Jan 15, 2007 4:50 pm
Location: Stockholm, Sweden
Contact:

Post by VilleK » Mon Feb 23, 2015 4:58 pm

Another topic I'm interested in is multiplayer/network support. Although possible to build from scratch using TCP/IP protocol, I'm sure there are libraries out there which simplify things and are cross-platform. Anyone know of such a library?

User avatar
Kjell
Posts: 1668
Joined: Sat Feb 23, 2008 11:15 pm

Post by Kjell » Mon Feb 23, 2015 5:46 pm

Ehm,
VilleK wrote:Anyone know of such a library?
Forgot about 39DLL? ;)

K

User avatar
Rado1
Posts: 770
Joined: Wed May 05, 2010 12:16 pm

Post by Rado1 » Mon Feb 23, 2015 6:27 pm

Carver413 wrote:there is also festival, flite and open mary tts that work in linux. using shell scripts you can create sound files. not as good as an api but it works. flite is my favorite at the moment its small and there are 11 voices you can get for it. open mary has some nice sounding female voices but can be a bit buggy at times.
I'm aware of these TTS systems/libraries but all of them seems to be large, or at least their voice files are large comparing to built-in Windows TTS. Anyway, Carver413 are you able to compile flite on Windows and maybe also on Android? I was not able, there are missing some files in src distribution for v.2.0.0.

User avatar
Rado1
Posts: 770
Joined: Wed May 05, 2010 12:16 pm

Post by Rado1 » Mon Feb 23, 2015 6:30 pm

Kjell wrote:Forgot about 39DLL? ;)
Unfortunately, 39DLLzge works only on Windows.

User avatar
Kjell
Posts: 1668
Joined: Sat Feb 23, 2008 11:15 pm

Post by Kjell » Mon Feb 23, 2015 6:38 pm

Hi Rado1,
Rado1 wrote:Unfortunately, 39DLLzge works only on Windows.
There are various ports ( 39DLL-4-Linux / 39dylib etc. ) available for Linux, Mac OS X, Android and iOS though :wink:

K

Post Reply