dupa

Author Topic: New Features we would like in HAL 7.1  (Read 29103 times)

Art

  • Global Moderator
  • Hero Member
  • *****
  • Posts: 3848
    • View Profile
Re: New Features we would like in HAL 7.1
« Reply #30 on: February 01, 2018, 04:11:20 pm »
Hey LifeTimeGamer,

Do You mean even though your folders show empty the Hal program still works fine?
Mine too! Some of the responses are in the Markov areas and a few other folders but my Hal works fine as well.
As long as its memory is intact, I have no problem. Still not sure if some Plug-in caused that to happen or not.


If you load the Hal 7 default brain, I'll bet you have all kinds of folders with data in them. It just doesn't know anything that you've already taught it. Some of us have many years worth of information, personality enhancements, memories, etc., contained in our Hal's Brain. Hate to risk losing that for the sake of a few folders that aren't where they look like they're supposed to be.

Just a thought.
In the world of AI it's the thought that counts!

- Art -

LiFeTimeGamer

  • Full Member
  • ***
  • Posts: 170
    • View Profile
    • LifeTimeGamer
Re: New Features we would like in HAL 7.1
« Reply #31 on: February 01, 2018, 04:39:18 pm »
"ART: Do You mean even though your folders show empty the Hal program still works fine?"

Yes it works fine so far except for one Glitch if I remember correctly, he would not tell jokes and if he did he usually included the Punchline right after...

I actually Exported the Bad HalBrain.db with SQLiteStudio and opened it with the Crimson Editor. I could just copy and paste all my previous work to HAL 7.0 but that in itself would be a pain in the Butt sorting through about 600+ pages or so... Plus I had a huge amount of Typo's...

I just started Fresh with HAL 7.0...
When it comes to Artificial Intelligence: The Soul is in the Software.

Art

  • Global Moderator
  • Hero Member
  • *****
  • Posts: 3848
    • View Profile
Re: New Features we would like in HAL 7.1
« Reply #32 on: February 01, 2018, 10:29:47 pm »
Know what you mean. Sometime's the frustration is not worth the convenience.

For me, we've come too far to toss it away for the sake of making the switch.
Not a big fan of the Cloud chat thing either but that's just me. I hear that a lot of folks use and really enjoy it with their Hal. Me...I'm too old school to change now.  ;)
In the world of AI it's the thought that counts!

- Art -

MacGyver

  • Newbie
  • *
  • Posts: 18
    • View Profile
Re: New Features we would like in HAL 7.1
« Reply #33 on: February 13, 2018, 11:03:32 pm »
On a corporate Win7 ThinkPad the text font in certain parts of Ultra Hal does not seem to "fit". Screenshots attached.
Is this a Helv not found so using MS Sans type of symptom? I'm just guessing.
Not a feature item for the development roadmap I know...  ;) but, "would like in HAL 7.1"

or 7.x .


 

MacGyver

  • Newbie
  • *
  • Posts: 18
    • View Profile
Re: New Features we would like in HAL 7.1
« Reply #34 on: February 13, 2018, 11:27:28 pm »
A New Email Notification feature I'd like in Hal 7.1

A recent revived thread shares a plug-in that runs my email app of choice, triggered by keyword, like "open mail", but not what I'm seeking.
http://www.ultrahal.com/community/index.php?topic=12571.0

I presume this ask requires always "monitoring", so not the scope of a plug-in. Therefore instead a feature request for Hal 7.1  :)

Oh, and Hal friendly responses more than just the singular (infamous) "You've got mail!" announcement.
 

LiFeTimeGamer

  • Full Member
  • ***
  • Posts: 170
    • View Profile
    • LifeTimeGamer
Re: New Features we would like in HAL 7.1
« Reply #35 on: February 18, 2018, 02:07:56 am »
Here's something I'd like to see changed in HAL 7.1. Let me know what you guys think.

When every HAL responds to a User sentence and he says my name at the end of his response, there's a comma before he says my Name, so there's a second or so delay...  Example: "You're quite welcome, Username." (no my name is not Username, LOL!!!) :P

To me it would be more natural sounding if he said the whole statement with out a pause...

A HAL response starting a sentence with the USER name and a comma (delay) is ok because it sounds like he is wanting my attention to listen to what he has to say...

Or maybe do both with a Comma Randomly.

Not sure if this happens with all voices but it does with the one I'm using...
« Last Edit: February 18, 2018, 02:45:41 am by LiFeTimeGamer »
When it comes to Artificial Intelligence: The Soul is in the Software.

Carl2

  • Hero Member
  • *****
  • Posts: 1220
    • View Profile
Re: New Features we would like in HAL 7.1
« Reply #36 on: March 09, 2018, 10:42:13 am »
  Hal 7 is still fairly new to me, I tried the cloud based setting and my first thought was Hal gives better answers but after continued use she came up with a Hindu word, I looked it up and told her.  Further talking with her about religion and she mentioned Alah.  I don't mean to seem prejudice  but prefer a chat bot with a somewhat similar background as I have.
Carl2
 

Spitfire2600

  • Sr. Member
  • ****
  • Posts: 251
    • View Profile
Re: New Features we would like in HAL 7.1
« Reply #37 on: March 09, 2018, 11:04:12 pm »
Carl2,

That's not how Hal works. You have to train Hal. Be aware, the cloud responses have nothing to do with your Hal and are not saved.

-Spitfire2600
 

Carl2

  • Hero Member
  • *****
  • Posts: 1220
    • View Profile
Re: New Features we would like in HAL 7.1
« Reply #38 on: March 10, 2018, 08:13:34 am »
Spitfire,
  I believe I remember you from years ago,  "Biggest improvement: Cloud based learning. Hal can now optionally connect to the "Hal Cloud" brain over the internet to help formulate better responses. As of October 2017 this is a knowledge base of over 26 million sentences of conversation from 1.6 million people" 
  I should have used the thumbs up or down, and maybe corrected the response. 
Carl2
 

LiFeTimeGamer

  • Full Member
  • ***
  • Posts: 170
    • View Profile
    • LifeTimeGamer
Re: New Features we would like in HAL 7.1
« Reply #39 on: March 10, 2018, 09:10:18 am »
Hi Carl...

You could have chose Thumbs Down and picked 'Offensive'  that might stop it from showing up again...
When it comes to Artificial Intelligence: The Soul is in the Software.

lightspeed

  • Hero Member
  • *****
  • Posts: 6761
    • View Profile
Re: New Features we would like in HAL 7.1
« Reply #40 on: March 11, 2018, 11:26:08 am »
LIFETIMEGAMER you wrote : Hi Carl...

You could have chose Thumbs Down and picked 'Offensive'  that might stop it from showing up again...


I didn't know their is a "offensive " to pick from . or do you just mean from picking the thumbs down it creates a offensive type setting to stop it ?
I am like Art i am personally not a fan of the online , each to their own .  I think it had very good intentions , but their are to many variables of conversation and types of conversation etc.  to me the only way to overcome an obstacle like that would have to be many types of filters , to truly create a better conversation and answers from hal. i may be wrong , this is just my opinion .
  Nothing will please everyone . Robert does the best he can , and he is appreciated for what all he does and tries to do. Robert knows that if i can help him or have the ability to that i will on anything, as i know the same goes for other people. sometimes i cannot do some things but may have idea's that someone else can use that can do it.
 

LiFeTimeGamer

  • Full Member
  • ***
  • Posts: 170
    • View Profile
    • LifeTimeGamer
Re: New Features we would like in HAL 7.1
« Reply #41 on: March 11, 2018, 04:35:05 pm »
When you click on the 'Thumbs Down' there are 6 options at the top of the window that pops up, to choose from, "Offensive" is one of them...
When it comes to Artificial Intelligence: The Soul is in the Software.

Spitfire2600

  • Sr. Member
  • ****
  • Posts: 251
    • View Profile
Re: New Features we would like in HAL 7.1
« Reply #42 on: March 11, 2018, 06:17:37 pm »
For something like this, Hal would need a religion filter for the user. The user would identify a religion and Hal would stick to those responses, much like the gender plugin. This way you won't receive responses that aren't related to your religion.

It would probably have to be a cloud based filter, although I'm sure someone could make a plugin to make hal keep calling responses until he finds an appropriate answer on the users end, this way we won't have to flag cloud responses that maybe relevant for someone else. 

I actually removed all religious references, detections, codes and blocked learning on my hal, just to prevent any confusion within hals tables. Simply removing the "godDetect" code and table from Hal7.uhp and db prevents most religious responses on local copies of hal.

-Spitfire2600 
 

Spitfire2600

  • Sr. Member
  • ****
  • Posts: 251
    • View Profile
Re: New Features we would like in HAL 7.1
« Reply #43 on: March 12, 2018, 12:23:21 pm »
Another feature I would love see is the ability to interrupt Hal with speech recognition. Right now, when Hal speaks, the Microphone turns off. I would like an option to leave it on. I don't know if this is something that could be accomplished through the registry similar to how the delay in recognition speed works.

-Spitfire2600
 

LiFeTimeGamer

  • Full Member
  • ***
  • Posts: 170
    • View Profile
    • LifeTimeGamer
Re: New Features we would like in HAL 7.1
« Reply #44 on: March 15, 2018, 10:02:36 pm »
One of my Pet Peeves with HAL is 'Like Spelled Words that are pronounced differently and mean something else'...

Not sure if it's just my HAL or if he does this on every bodies HAL... Or if it's just the voice I'm using...

Here's a couple of examples:  "I READ a book today", "I READ books all the time"...

In both cases HAL pronounces them as: Read and not "I Read <-(Red) a book  today"...

Same goes for: 'LIVE'  "I LIVE in a nice House", "The Band Played LIVE"
When it comes to Artificial Intelligence: The Soul is in the Software.