Announcement

Collapse
No announcement yet.

Want to replace Astra with my own voice

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Want to replace Astra with my own voice

    I've added Brian from the Ivona collection as my default TTS profile in Windows. He sounds very much like Kevan Brighting from the Stanley Parable. But when I go to replace Astra's sounds in Voiceattack (such as "deploy scoop") I find
    Execute command, '((Cargo Scoop)) (and wait until it completes)
    Execute command, '((RS-Cargo Scoop0))' (and wait until it completes)
    No mention of the sound file being used, but rather commands being used. . I didn't even know about commands in Voiceattack until I started using your profiles. I see a number of commands in the 1.6.5 Basic Profile.vap file but how would I go about removing the Astra sound references without losing multipart actions? Thanks.

  • #2
    The sound files that are referenced are actually inside the command for ((RS-Cargo Scoop)), so if you look at THAT command, then you will see the sound files that we reference.

    Now here is the long part for you.

    All of the sounds that you hear in the ASTRA voice pack (and all of our voice packs to be honest) are pre-recorded responses. So this means that you would have to change each and every single response, from just being a sound file being played to a "Say Something with TTS" action. And then type out the contents of each response, on a entry by entry basis.

    Just as a piece of information: ASTRA has roughly 2800 individual speech files.
    1) I reject your reality.... and substitute my own
    2) Not to be used when upset... will void warranty
    3) Stoke me a clipper i will be back for dinner
    4) Never tell Gangrel to do anything... he will probably get it wrong
    WARNING! Swedish wall-of-text hits you for bork-bork-bork damage!

    Comment


    • #3
      LOL That does sound like a lot and obviously takes a lot of work. Spent hours myself creating just 9 responses for deploying and retracting the cargo scoop. Using NaturalReader to create the sound files and it's tricky creating inflections with just TSS. Anyway, I'm very happy with my results. http://www.mediafire.com/download/b3...coopVoices.rar

      So instead of trying to alter the cmd files I just removed the references to them and replaced them with my own. I noticed that, instead of key assignments, some of your cmds have on/off switches i.e. "0" and "1". I've barely gotten started on this game so unsure how complex the actions are in your cmd assignments for some things. Also, unclear why so many keyboard cmds are disabled in your profile.

      Anyway, off to play with Audacity to see if I can simulate better even better inflections. Some day I'll get around to actually playing Elite Dangerous.

      Comment


      • #4
        Originally posted by GlutenFreeSpaceDandy View Post
        LOL That does sound like a lot and obviously takes a lot of work. Spent hours myself creating just 9 responses for deploying and retracting the cargo scoop. Using NaturalReader to create the sound files and it's tricky creating inflections with just TSS. Anyway, I'm very happy with my results. http://www.mediafire.com/download/b3...coopVoices.rar
        Welcome to voice pack work

        So instead of trying to alter the cmd files I just removed the references to them and replaced them with my own. I noticed that, instead of key assignments, some of your cmds have on/off switches i.e. "0" and "1". I've barely gotten started on this game so unsure how complex the actions are in your cmd assignments for some things.
        Personally I would actually just edit the ((RS-Cargo scoop)) command itself. This keeps the amount of "failure" possible down to a minimum. If you want to edit responses to commands, then just edit the ((RS-XXXXX)) commands. Nothing will break *mechanically* with the command. Just that if something does go wrong, it won't say anything (which is a good way to test the system when you think about it).

        We use commands with various switches for interesting stuff such as being able to give appropriate responses to "Deploy cargo scoop", both when it is closed (so that it opens, and it goes "opening hatch"), and when it is open (so that it goes "hatch is already open", and does nothing).

        In the basic profile, doing the deploy cargo scoop, would just toggle the status of the scoop, so if it was open, it would close. If it was closed it would open. And that is with a command telling the computer to *open* the hatch.

        We use other switches so that the docking command knows how many tabs to move left/right and then request docking. This kind of stuff works *really well* for the 100% voice controlled setup. But as soon as you start fiddling with tabs with the joystick/keyboard directly, it can get a bit messy and start breaking[1]. This is why we recommend the beta profile for those who are experienced with both VoiceAttack AND Elite Dangerous.

        The basic profile doesn't use toggles/variables etc to the degree of the beta profiles. But then again, it isn't as "automated" as the beta profile either.

        Also, unclear why so many keyboard cmds are disabled in your profile.
        This is because those commands form part of other commands. We don't want these to be activated *directly* by their command name, but by other phrases instead. Keybinds are an example of this. A person only needs to check 60 (ish) keybind commands to make sure that they match up with what is in Elite Dangerous. Much better than checking over 1000 individual commands....

        We use the ((commands)) (disabled) as building blocks to make up larger more complex sections of the profile. So being able to activate them by name *directly* wouldn't allow us to do this as well (if at all)

        As an example. I have a "random profile" switching command[2]. All this does is after a random period of time, it swaps my voice attack profile to another random one. If I didn't have the ability to use ((commands)) then it would be very hard for me to do this. And to keep it running constantly in the background.

        However, with a ((random time)) command, and a ((Random profile)) command that I have made, it is *very* easy to do. And to add in new wait period times/profiles just means I have to edit two different "base" commands, instead of quite possibly 10 or 11 individual profiles[3]

        [1] There are ways around this, but it is something that the person would have to setup themselves to do so.
        [2] This is in my personal "profile switch" profile. Really handy for when you have a lot of voice packs. Being able to change voice packs according to mood with just your voice is fun.
        [3] The wonders of the "include commands from X profile" option allows this to work *marvelously*
        1) I reject your reality.... and substitute my own
        2) Not to be used when upset... will void warranty
        3) Stoke me a clipper i will be back for dinner
        4) Never tell Gangrel to do anything... he will probably get it wrong
        WARNING! Swedish wall-of-text hits you for bork-bork-bork damage!

        Comment

        Working...
        X