Release EDDI 3.3 - Bring your cockpit to life

[VerticalBlank]

Optional update EDDI v 2.4.3 released.

• EDDI will no longer nag users to send logs for commodity definition errors (and there was much rejoicing).
• Amended the age calculations for the fact that age is reported in millions of years, not years.
• Amended reporting of stars less than a million years old.
• Amended reporting of Herbig-Haro objects: they are now recognised.
• Enhanced the reporting of Wolf-Rayet stars.
• Sundry punctuation tweaks to make the speech more natural.
• 'Entered signal source' thoroughly re-written to better report both human and Thargoid signal sources.
• Worked around non-compliance of CereProc voices with industry standards that would cause EDDI to revert to a system default voice.
• Many other sundry fixes, see the full change log.

 

[Tkael]

We're aware of an issue with stellar age being reported incorrectly after a scan. A fix is already on the way.

 

[Brigetiol1]

Hi,

First of all, thanks for the update! The new functions sound very interesting : emphasize, speedpitch, etc... unfortunately, they don't work on my PC, so is behaving Play() too
I've tried the examples from the SpeechResponder help documentation, no effect. It is like SSML is disabled somewhere deep in my PC for EDDI. In all cases, when I've tried these functions to test, there was no error message, just no effect at all.

I have Win7, VoiceAttack, Ivona Kimberley voicepack, in EDDI the 'disable phonetic speech' box is UNchecked. The G19 Companion App also uses these SSML tricks, they work there.

Could you give me some hint, how to make all these work?

Thanks in advance!

 

[BumbleB]

Small thing I'm trying to swap personalities on the fly using "setspeechresponderpersonality" voice attack command but it doesn't seem to change the personality.

Excuting External Plugin "EDDI 2.4.3"

Plugin Context "setspeechresponderpersonality"

Text Variables: "Infinity 2.0"

Wait for the plugin function to finish before continuing *ticked*

*from the wiki*

setspeechresponderpersonality


This function changes the speech responder's personality. The name of the personality should be a text variable with the name 'Personality'.

Then changing to another profile in Voice Attack profile in voice attack.

I probably doing something simple wrong but I can't see it atm.

Also is there a way for me to get which personality is currently loaded as a variable? So I can just ask Voice attack which personality is currently loaded.

Big thanks for any help

 

[J. Calvert (Joshua)]

Did you set the text variable Personality to “Infinity 2.0” before calling the plugin ?

 

[tulextreme]

AWESOME job cmdr Darkcyde! I was very anxious for this release! Just now ima gonna start to mingle with this great 'personality' ! Thank you so much for sharing your gr8 work with us cmdr!

 

[Tkael]

Hi,

First of all, thanks for the update! The new functions sound very interesting : emphasize, speedpitch, etc... unfortunately, they don't work on my PC, so is behaving Play() too
I've tried the examples from the SpeechResponder help documentation, no effect. It is like SSML is disabled somewhere deep in my PC for EDDI. In all cases, when I've tried these functions to test, there was no error message, just no effect at all.

I have Win7, VoiceAttack, Ivona Kimberley voicepack, in EDDI the 'disable phonetic speech' box is UNchecked. The G19 Companion App also uses these SSML tricks, they work there.

Could you give me some hint, how to make all these work?

Thanks in advance!

These new functions rely on the compliance of your installed voice to documented SSML standards so your mileage may vary. Cereproc voices, for example, are not particularly good with SSML compliance.
If phonetic speech is enabled and you have a compliant voice then you should be able to use these new functions.

 

[Tkael]

Small thing I'm trying to swap personalities on the fly using "setspeechresponderpersonality" voice attack command but it doesn't seem to change the personality.

*from the wiki*
The name of the personality should be a text variable with the name 'Personality'.

Then changing to another profile in Voice Attack profile in voice attack.

I probably doing something simple wrong but I can't see it atm.

Also is there a way for me to get which personality is currently loaded as a variable? So I can just ask Voice attack which personality is currently loaded.

Big thanks for any help

Did you set the text variable Personality to “Infinity 2.0” before calling the plugin ?

What J. Calvert said. The text variable should be named 'Personality', and it should set to a value equal to the actual name of your personality (so the variable 'Personality' contains a value of “Infinity 2.0”) before the plugin is called.

 

[Tkael]

Optional update EDDI v 2.4.4 released.

• Fixed a bug that was causing some SSML related functions (e.g. Pause()) to not render correctly.
  
  
• Fixed unit conversion of the star's age in star scans. They should no longer report every star as "one of the oldest".

 

[J. Calvert (Joshua)]

Hi,

First of all, thanks for the update! The new functions sound very interesting : emphasize, speedpitch, etc... unfortunately, they don't work on my PC, so is behaving Play() too
I've tried the examples from the SpeechResponder help documentation, no effect. It is like SSML is disabled somewhere deep in my PC for EDDI. In all cases, when I've tried these functions to test, there was no error message, just no effect at all.

I have Win7, VoiceAttack, Ivona Kimberley voicepack, in EDDI the 'disable phonetic speech' box is UNchecked. The G19 Companion App also uses these SSML tricks, they work there.

Could you give me some hint, how to make all these work?

Thanks in advance!

Pause() also wasn't working but it's fixed in 2.4.4 so maybe this is fixed too.

 

[Tkael]

Pause() also wasn't working but it's fixed in 2.4.4 so maybe this is fixed too.

Yes, that should also make a difference.

 

[BumbleB]

What J. Calvert said. The text variable should be named 'Personality', and it should set to a value equal to the actual name of your personality (so the variable 'Personality' contains a value of “Infinity 2.0”) before the plugin is called.

AH! *penny drops*

You answered both of my question in one. Knew I wasn't looking at it right.

Set Text [Personality] to 'Infinity 2.0'
Execute external plugin, 'EDDI 2.4.4' and wait for return

 

[T1-Jafo]

Umm, are you sure? When I was using Windows 7 Pro I was able to use all 3rd party voices.



It is a buyer decision, but as a developer I can say that the CereProc voices have always required workarounds and still do, because not only do they not meet industry standards (such as support for the PLS+XML lexicon format and the IPA alphabet), but they obstruct supporting industry standards by crapping out when valid data (such as the xml:lang tag) is supplied.

jgm also said as much in the old thread. The Ivona voices are in my experience much more compliant with industry standards.

It is a shame that the text-to-speech market is so lax about meeting documented standards. We will endeavour to support all vendors for as long as is feasible, but make no mistake, if push comes to shove we will focus our limited time on supporting voices that are standards-compliant.

As a developer just trying to maintain a standards-compliant text-to-speech app, my personal opinion of CereProc is that I would like them to DIE IN THE FIRE OF A MILLION SUNS. Just sayin'

I wish I had known about the trouble developers are having trying to get the folks at CereProc to work with everyone before I made a purchase from them.

Yes, I am certain. Since I am running Windows 7 64 my choices are limited and I am forced to look at 3rd party voices. You can download and install Microsoft Speech Server v11 but in order for it to work properly in Windows 7 64, you need to do a couple of simple registry modifications. (Basically change the registry folder path of the TTS tokens in two locations). It has something to do with the way Windows 7 64, (and 2008 Server R2), requires various entries in the Speech Token section of HKLM.

Purchasing a 3rd party voice for $34 is an easy alternative if folks don't want to mess around changing the registry. *EDIT* - I just searched, and although I missed it the other day, end users can also purchase voices from Ivona that work with other SAPI5 applications for $45. http://www.textaloud.com/englishbritish.shtml

But I am just an end user and have no idea (until recently), what SAPi (x) was.

Great work you've done with EDDI 2.x, btw. It really helps with immersion.

 

[Brigetiol1]

Hi,

First of all, thanks for the update! The new functions sound very interesting : emphasize, speedpitch, etc... unfortunately, they don't work on my PC, so is behaving Play() too
I've tried the examples from the SpeechResponder help documentation, no effect. It is like SSML is disabled somewhere deep in my PC for EDDI. In all cases, when I've tried these functions to test, there was no error message, just no effect at all.

I have Win7, VoiceAttack, Ivona Kimberley voicepack, in EDDI the 'disable phonetic speech' box is UNchecked. The G19 Companion App also uses these SSML tricks, they work there.

Could you give me some hint, how to make all these work?

Thanks in advance!

Hi,

thanks for the quick fix, Play() is working at last, I was waiting for this function for long time !

Most of the SSML functions work, only Emphasize() seems not to be working, but the rest gives me also lots of possibilities to play around my profile!

Best regards!

 

[Tkael]

Hi,

thanks for the quick fix, Play() is working at last, I was waiting for this function for long time !

Most of the SSML functions work, only Emphasize() seems not to be working, but the rest gives me also lots of possibilities to play around my profile!

Best regards!

As I said before, your mileage may vary.

With Ivona Amy, I can't detect any change in the speech pattern from the Emphasize() function (thought the others work nicely). But with Microsoft Hazel, I can hear a definite difference when using the Emphasize() function.
In spite of the variation from voice to voice, I'm keen to see what people do with these new speech functions.

 

[Leo Sagan]

Hello VerticalBlank,

Thank you so much for your hard work, EDDI seems great.

I did not install it yet because I use an heavily tweaked HCS voice pack (ASTRA) and I was worried that it would cause troubles.

Is it possible / difficult / useful to have both ? How should I install it ?

Thanks in advance

 

[Shadow Doctor K]

New Variable Request

Hey guys there are so many different fourms for this plugin, and the getgub seems to have low activity. If this is the wrong place to do this please let me know and I will correct my post.

So I was wondering if we could get a new variable added for voice attack to use. The ships loadout event has everything but one thing, engineered status. So using an example from my log below:

{ "timestamp":"2017-11-07T11:12:49Z", "event":"Loadout", "Ship":"Federation_Corvette", "ShipID":11, "ShipName":"", "ShipIdent":"", "Modules":[ { "Slot":"HugeHardpoint1", "Item":"Hpt_BeamLaser_Gimbal_Huge", "On":true, "Priority":0, "Health":1.000000, "Value":8746160, "EngineerBlueprint":"Weapon_Efficient", "EngineerLevel":5 }

I was wondering if the two variable could be for the EngineerBlueprint, and EngineerLevel. I would like to use those to trigger an action in a script i'm writting but this the only way I believe I can detect the engineering status of a weapon. If there is another way I'd love to know I haven't been able to find a way to detect if a module has been engineered or not.

Thanks,
CMDR Shadow Doctor K

 

[Shadow Doctor K]

New Event Request

Sorry, I'm not able to modify my previous post. Since i'm a new member all my post have to be approved. So its not there for me to edit.

I was also wondering how hard it would be to add these as events with a few new variables?

"event":"ModuleStore", "Slot":"Slot04_Size6", "StoredItem":"$int_fighterbay_size6_class1_name;", "StoredItem_Localised":"Fighter Hangar"
"event":"ModuleRetrieve", "Slot":"Slot04_Size6", "RetrievedItem":"$int_fighterbay_size6_class1_name;", "RetrievedItem_Localised":"Fighter Hangar"
"event":"ModuleSell", "Slot":"Slot02_Size7", "SellItem":"$int_cargorack_size7_class1_name;", "SellItem_Localised":"Cargo Rack", "SellPrice":942736,
"event":"ModuleBuy", "Slot":"Slot02_Size7", "BuyItem":"$int_cargorack_size7_class1_name;", "BuyItem_Localised":"Cargo Rack", "BuyPrice":942736,

I'm looking to update the load out of the ship everytime a module change is detected, right now the only event trigger to help with this is the Ships Loadout event.

CMDR Shadow Doctor K

 

[VerticalBlank]

Hello VerticalBlank,

Thank you so much for your hard work, EDDI seems great.

I did not install it yet because I use an heavily tweaked HCS voice pack (ASTRA) and I was worried that it would cause troubles.

Is it possible / difficult / useful to have both ? How should I install it ?

Thanks in advance

I'm the representative standalone user on the dev team but I'm sure that VoiceAttack users T'kael and Hoodathunk will chime in if I get anything wromg.

It is entirely straightforward: tweaks to your HCS voice pack shouldn't be an issue. If VoiceAttack is in the default location "C:\Program Files (x86)\VoiceAttack" then the installer will do the right thing. Otherwise, please see https://github.com/EDCD/EDDI/wiki/VoiceAttack-Integration.

If you have any problems please ask.

 

[VerticalBlank]

Hey guys there are so many different fourms for this plugin, and the getgub seems to have low activity. If this is the wrong place to do this please let me know and I will correct my post.

So I was wondering if we could get a new variable added for voice attack to use. The ships loadout event has everything but one thing, engineered status. So using an example from my log below:

{ "timestamp":"2017-11-07T11:12:49Z", "event":"Loadout", "Ship":"Federation_Corvette", "ShipID":11, "ShipName":"", "ShipIdent":"", "Modules":[ { "Slot":"HugeHardpoint1", "Item":"Hpt_BeamLaser_Gimbal_Huge", "On":true, "Priority":0, "Health":1.000000, "Value":8746160, "EngineerBlueprint":"Weapon_Efficient", "EngineerLevel":5 }

I was wondering if the two variable could be for the EngineerBlueprint, and EngineerLevel. I would like to use those to trigger an action in a script i'm writting but this the only way I believe I can detect the engineering status of a weapon. If there is another way I'd love to know I haven't been able to find a way to detect if a module has been engineered or not.

Thanks,
CMDR Shadow Doctor K

Hi,

If "getgub" is an auto-correct for "GitHub" then please check that you have the correct URL https://github.com/EDCD/EDDI. The activity there is definitely not low, so maybe you are looking at an older repo or a fork.

GitHub is not for everyone. The alternatives are this thread and the EDDI channel in the EDCD Discord.

The current build 2.4.4 of EDDI should be tracking your engineering correctly. The best way to test that is currently via export to EDShipyard.

If you have a reproducible case where EDDI isn't tracking your engineering correctly, we want to hear about it!

o7