regolovr
patreon
Regolo Engine S 1.50
🕑 Added 2022-06-09 21:48:23 +0000 UTCOne week? Who said one week 😂 ?
I'm joking, I know, I'm a bunch of days late, forgive me 🙏 .
Anyway, despite the delay the second Regolo Engine S installment is here with a lot of new amazing stuff.
Let's go check them.
Some important notices before going on
1. This is a beta version that uses external software and (optionally) connects to Microsoft cloud platform where you could be billed for services used. Please consider to check Azure cognitive services free/billing services terms and conditions before using Regolo Engine S with it.
Regolo Engine software and developer are not accountable for any billing issue or unexpected payment arose as consequence of its usage.
This is an unfinished product and a beta version and doesn't contain any mechanism that can prevent Microsoft from charging you with any amount of money.
All the details I know and can deliver to support are in the related paragraph.
2. Also, Regolo Engine S doesn't contain all the legacy stuff published before Simulation mode. If you want to try them you can download Regolo Engine 1.30. It's the last release that supports them.
3. There is an official doc that was updated till Regolo Engine S release. So it doesn't contain newest information as the Regolo Talking Machine or connection to the Azure platform. A lot of information are valid though, as the plugin installation ones.
Here' the link: Regolo Engine doc
The video
What's new
Here's an overview of the new stuff, check the next paragraphs for insights and instructions on some of them.
Simulation 1.50
- real time translation in many different languages available through Microsoft Azure
- additional languages and female/male voices available
- custom voice pitch, rate and volume
- objects management improved
- objects can be used in different ways and contexts (sitting, laying, dancing, interacting with boy...)
- girl can express herself about objects
- QoL fixes (e.g.: boy not twisting his torso anymore, less sudden corruptions overall that interrupt experience,...)
UI
- you can now choose the clock position among two (in the girl ui setting panel): "Up left" and "Bottom right"
misc
- you can set initial clock position in the config file
- you can set initial RTM language in config file
- you can set initial RTM voice in config file
- you can set initial RTM pitch in config file
- you can set initial RTM rate in config file
- you can set initial RTM volume in config file
- girl now reaches more easily some tricky destination (i.e.: objects placed on other furniture)
- boy's feet now follow movements without delay
- girl now uses hands to explore environment while walking (like she does while dancing)
- girl can use object she's dragging around to explore environment while walking and dancing
- extroversion's default value is now 0.5 (chat/speech every 50 secs)
- general optimization for the scheduler (it runs faster)
audio (enchanter)
- Regolo Talking Machine 1.1 offers modulation and translation capabilities
- more than 10 different languages
- about 40 different voices to choose from (both female and male)
fixes
- located and eliminated cause of boy twisting his torso after bad collision or laying down
- because of the above laying down with the boy has become more easy and smooth
- located the cause of many corruptions overall that were interrupting experience (many of them while girl was laying on the bed)
- girl had issues detecting obstacles in some occasions, this has improved
- girl got stuck on some occasions because of wrong height calculations. This has been solved
current known bugs
- still mix of old and new sentences (i.e.: found an object, hanging around...)
- footjob while laying down doesn't work well, investigating
- cumshot is a lot unstable during some occasion and gives issue
- licking dildo creates bad movements on some occasions (like when she is laying down)
- more in general testing old stuff for object integration I realized there are still a lot of bugs and issues. I revised some but objectively there are still a lot to deal with
As always, if you notice something that is not working fine please let me know and I will try to fix it asap.
Luke, tell your uncle if he gets a translator
So 1.50 introduces a new amazing feature, the real time translation.
This means that every sentence build in English by the engine can be immediately translated and chatted/pronounced by the girl using the new language.
I chose to add it since Azure offers this service among the ones with a free monthly threshold, that in this case is 2 millions chars of any combination of standard translation and custom training free per month.
In order to enable it you have to create the service on the Azure platform. Below you will find a guide about the translator service. You can find the general and text to speech ones in the 1.40 release post.
After creation you will be able to retrieve key1 and region code. To enable it you have to put those codes in the plugin config file ("translatorKey1" and "translatorRegion" fields). "translatorEnabled" must be true too.
RegoloTalkingMachine executable should be running before the plugin is started in order to have the translation (and the speech synthesis as well, check 1.40 release post).
Here's a detailed step to step guide to enable translation service on Azure platform. Please refer to the 1.40 guide to enable text to speech or to create an Azure Platform account.
AZURE Translator step by step guide
- 1. access to Azure home page: https://portal.azure.com/#home
- 2. on the vertical drop down menu on the upper left corner click on "Create a resource"
- 3. search for "Translator" in the search field, select it and then click on "Create" on the next page (the one with the service details)
- 4. on the creation editor you need to put
* the subscription (the current one is already selected)
* the resource group (use can use text to speech one if you want)
* the location (es: West US)
* the name
* the pricing tier (choose Free F0) <-------------- very important - 5. click on validate and create, and then on create again, distribution will start and text to speech resource will be created
- 6. gratz, your translator resource has been created, now you can go to the resource page (from the home, the dashboard, from the vertical menu or from the blue button appeared after creation "go to resource")
- 7. on the resource page, click on "keys and endpoint" on the left menu
- 8. copy key1 and location and put them between quotes in the related translator fields (translatorKey1 and translatorRegion) of the "talkingMachine" section in the regoloEngine_E.json config file. Also don't forget to put the "translatorEnabled" field to "true"
As for the text to speech service Azure will require you to upgrade your account after a month. This anyway doesn't take away your monthly free quota but makes you pay if you cross the threshold.
On this regard I discovered that you can check a log of the speech synthesis activity on Azure though. This is true for translator service as well I think.
To do that you can go on Metrics on the left vertical menu, then select you cognitive speech service and then Sythesized Characters. It will show a graph of the characters sythesized in the selecte period (upper right corner).
That's a lovely, lovely voice
Yes, neural synthesis gives us a lot of options to find our preferred voice and intonation.
Using the menus in the plugin general menu you can infact set a new language, a new voice (choose one that matches the language even if that's not a rule), and modify the pitch, the rate and the volume in real time!
You can use a voice with a different language as well to generate fun pairings and basically simulate accents.
I added male voices too, because you know, creativity.
You can even set their initial values in the config files, to save time.
If you activated the text to speech service you can always change voice, pitch, rate and volume, but if you want translation you have to activate the Azure translation service as explained in the previous paragraph.
So consider that if you haven't set the key + region for the translator service and you're not using the talking machine you will continue listen to English sentences.
Also there are some quality differences between different voices. en-US voices seem slightly more advanced compared to other ones, in my opinion, and produce better voice inflections.
And it's full of leathery objects...
I introduced object management a while ago, but it was very limited and basically a prototype. Now the girl can bring along any small object found wandering around and use it actively while interacting or doing a specific action.
First thing to know is that the girl will try to grab an object if it is red (both controls off), not excessively far and not too big. She will approach it and try to pick it up, regardless of its position.
If the object is not small she will analyze it and then throw it away. Otherwise she will take it with her and try to reach the boy to show it off. She will also comment about it (raise the extroversion bar if you want her to speak more often).
This means that she will maintain the grab and will manage the object during almost all possible interactions:
- walking
- sitting
- laying down
- dancing
- lewd status
- submissive status
She will not enter into exploring or climbing for now while grabbing an object, but I am thinking about adding them as well.
While grabbing it she can use the object to caress herself, or use it sexually if the object is a dildo. All of the above depends on disposition level as usual.
She can also manage the change of possession, so if you steal her the object she will comment and try to have it back. If that's a dildo you can use it near her face/mouth to have some reaction.
She will try to reach a near object even from those stances as well, so if you put it near her she will try to grab it. If she is laying down, for example, she will hold out her hand to reach it.
If the object becomes green (controls on) she will ignore it.
Otherwise if she's not doing anything special (read: while walking) she will throw the object away or place it on a table after a while (depending on perseverance).
Also, she can now use the object to interact with environment while dancing or walking.
Last note: don't use any cube!! They crash the girl every time (Cube and ISCube) and break the scene. It's possible to reproduce the issue even without plugin running, so it seems VaM thing.
Just try to place a cube on the ground, scale it down to about 0.3, set controller to red and make the girl touch the top surface with her hands...
Lots of fun uh 😂 . This single thing stole me two days, yay...
Right now, I'm in the quality of life business
I told you I was focusing on QoL this release, so I picked up a couple of unresolved issues and I focused on them. Luckily I was able to understand the causes and now we have a smoother experience. I should thank you guys, I guess 👍 .
Anyway, the boy is not twisting his torso anymore after bad collision and when laying down. As I said it was a VaM issue but I found a workaround.
Since this has been fixed you can lay down with the guy more more easily now and be less worried about his movements.
I worked on the girl reset because of corrupted collisions as well. This triggered too often VaM restore mechanism that disabled girls' collisions and stuff. This seems solved now.
Also, a lot has been added during last months so I spent some times optimizing some critical parts that were suffering. I think now it's a lot more rare to see body part lagging too much, so it's a general improvement that benefits the overall experience.
Walking is better as well and girl gets stuck a lot less wandering around. Height management gave some problems but now they should be solved.
Obstacle avoiding also now is less picky and hands are basically free to explore while walking like they are while dancing. I think this adds a lot.
Fixed some minor things too, like running during a slow turn (very ugly).
Anyway, try out and let me know if you observe any improvement as well.
In the end
This was really a fun release, even if many things made me lose time and accumulate delay. I hope you will appreciate the new stuff though.
There are only about 20 days till the end of the month, and I think I know how to use them. I hope this will give me a way to deliver a light (but meaningful) release before the end of the month.
If this will not happen I will write the usual post full of apologies and stuff but I am confident enough it will not happen.
Then we have the July release, and I anticipate that it could be massive. Or better, I will try build a major release, but I suspect it will be very hard to make it work, so it's still a big question mark.
Anyway, we have a plan, let's see how it unfolds.
In the meanwhile, have fun and see you in a few weeks 🙌 .
RegoloVR
