Alright, I looked at the tenpenny thing. It is directional, and does stop when you pause, like mine should. -_- The only different between the setup at all, is that they run their back and forth off of a script variable, raising it with each line, while my uses a series of 'Begin SayToDone' blocks. That should not be affecting its sound output or anything... Also they use .ogg and I am using .wav, until I make the real recording which will be ogg. But that also should not make a drops difference in the 2d/3d part or game/menumode of it playing. Why is it doing this to me!!!?!?!?!
This is how I did mine:
1. Created a Talking Activator object for the intercom and placed a simple script on it:
scn Quest04S060IntercommSCRIPTbegin OnActivate ; Intercomm only works at stage 56 & 57 if( GetStage Quest04 > 55 && GetStage Quest04 < 58 && isActionRef player == 1 ) setTalkingActivatorActor JackFixitManRef if( GetStage Quest04 == 56 ) SetObjectiveCompleted Quest04 57 1 endif Activate endifend
Note that the intercomm governs the conditions under which it will respond to the player (or NPC), in my case Quest04 has to be between stages 55 and 58, and the actor has to be the Player. You can do something similar with yours. I also checked the "Quest Item" box in my activator. Voice type I used was "Maleadvlt01Default".
2. Use the Topics not Conversations. You DO need a GREETING topic to start things off, the rest can be normal Topics for the the NPC to say through the intercomm to the player. All Topics for this should Not be "top-level", though the GREETING may or may not have the "Say Once" flag or other flags checked.. For conditions in the first initial GREETING topic, you should use at Least 2:
2a. Make sure the right NPC is set (FunctionName: GetIsID, in my case the info is: NPC: JackFixitMan == 1.
2b. Also make sure you set the right quest stage condition so the intercomm Only responds to the player when the quest is in the right state. It may seem redundant as you also put conditions on the Talking Activator - both are actually required to cover all conditions and to make sure the right NPC talks during the right conditions. In this case my condition was FunctionName: GetStage, info: Quest: 'Quest04' == 56 (for quest stage 56 which is required for this NPC to speak the topic through the intercomm at you).
3. Now you can set-up additional topics if you want the conversation to carry on for awhile, use standard topic conditions at this point - once the NPC is talking to the player via the intercomm, you can chain the topics together normally. The Talking Activator has the conditions under which the entire conversation works (note my conditions in #1 above allow talk between a range of quest stages, then the topics control what is said, when).
4. Of course you will need a Real NPC for this, even if its an NPC in voice only that is never seen in-game, you need to create an NPC and have that NPC assigned to the dialogue/topics/etc.
5. Once you have the conversation flowing correctly from front to back, then it's time to worry about adding real audio to the voice. You can use .Wav files just fine, .ogg or .wav are both recognized by the GECK and the game, I think the .ogg gives you more compression but thats really not an issue with most Modding dialogue. If you run into voicing problems at this point, it has nothing to do with the activator or the dialogue - and you can focus on the sound-aspects to narrow the troubleshooting down.
Hope this helps, though I can't explain the .ogg issue, I have made about a dozen talking activators so far using this method.
Luck!
Miax