FD, please consider giving NPCs a voice

One of the best recent additions to the game (IMO anyway) was audio messages from station traffic control, giving much needed life to the game. It would be great if more NPCs got a voice.

Imagine you're doing another cargo run to some distant star somewhere. You get bored. Netflix is becoming an option to you.

Suddenly, a tone is dispatched through your speakers to alert you of an incoming hail, and a voice laced with menacing intent echoes throughout your bridge:

"Commander, disengage supercriuse operations and stand ready to contribute donations. This is your only warning. You run, you die. I await your compliance."

A bit better then bog standard text messages like "big haul blah blah" or "Die!" during combat, no? With ED's crack audio team, I'm sure they could pull this off, there's great variety of messages for the station crew so hopefully the same would be true for NPCs.
 
Eh? Yeah, I really want to listen to "Thats the one I'm after, the one with the big haul" or whatever nonsense the damn idiot npc halfwit keeps saying to me. With an empty hold.

As long as we can mute them, then fine. But, I warned you. Careful for what you wish for. Although it may make murdering him all the more palatable.
 
Eh? Yeah, I really want to listen to "Thats the one I'm after, the one with the big haul" or whatever nonsense the damn idiot npc halfwit keeps saying to me. With an empty hold.

As long as we can mute them, then fine. But, I warned you. Careful for what you wish for. Although it may make murdering him all the more palatable.

Oh yeah, this would be optional.
 
I'd be perfectly happy with indirect mostly indistinguishable radio chatter.

With little effort, it can be used to great effect.

Basics; you get more chatter in busier areas, docking and patrol related at docks, general banter out in supercruise.
USS specific chatter as you lock on or approach.

Then eerie silence out of the bubble.
Or not so silent, if something odd is going on.

FD have an excellent audio team, I'm sure they could do this!

(And direct audio would be amazing too!)
 
Because NPC communications are now recorded to the Journal e.g...

Code:
{ "timestamp":"2017-01-16T19:28:23Z", "event":"ReceiveText", "From":"$npc_name_decorate:#name=Chris Armstrong;",
"From_Localised":"Chris Armstrong", "Message":"$BountyHunter_HunterHostileSC_Relevant03;",
"Message_Localised":"There you are! That bounty makes you today's meal ticket.", "Channel":"npc" }


...it should be possible to write a program to parse the file in real time, extract the actual dialogue and send it to a text-to-speech engine. If you wanted to be clever you could compare each NPC name with a database of gender-weighted names and allocate a male or female voice accordingly. If you wanted to be really smart you could use a hash of the NPC's name as a seed for picking from a selection of voices, thus ensuring that the "same" NPC had the same voice each time you encountered them.

This is well beyond my meagre string-slicing skills but it seems like an ideal project for an enthusiastic coder to give it a shot if FD don't want to. Hell, if someone's already done it please let me know; I would love such a feature and a donation from my PayPal account stands ready... ;)
 
i like the idea - as long as the're enough variety in the chatter - im thinking hundreds or thousands of one liners. and that it keeps evolving (heck have a contest where community contributes to one-liners and vote to choose 10 or 15 every few months) - not like the same SC music we've been hearing since the Alpha days.
 
Was working on this for an EDDI script. EDDI currently won't allow access to the npc name and chat script, but if it did you could have your wish of voiced npcs. It would be possible to code in EDDI scripts all the cleverness needed to stop the npc voices becoming to repeative and anoying.
 
Eh, Star Citizen has fully voiced pirates and it gets old fast. I'd rather they spend the time and effort on voice acting for special NPCs like engineers or heads of powers, or others that move the story forward.
 
Last edited:
I always loved the AI chatter in Falcon 4. Would love it here for space to feel more alive - with one big proviso... We don't get the random 'open mic' chat spam we do now, rather, aside from direct communication addressed to the player, the AI comms. are limited to things like callsigns, vectors and destinations, as well as far more terse 'flavour' text.
 
Until they can get procedurally generated conversations sorted and couple it with a decent voice synthesiser, no to voices. It gets tedious enough listening to SLF pilot and seeing the local pirates announcing their intentions.
 
One of the best recent additions to the game (IMO anyway) was audio messages from station traffic control, giving much needed life to the game. It would be great if more NPCs got a voice.

Imagine you're doing another cargo run to some distant star somewhere. You get bored. Netflix is becoming an option to you.

Suddenly, a tone is dispatched through your speakers to alert you of an incoming hail, and a voice laced with menacing intent echoes throughout your bridge:

"Commander, disengage supercriuse operations and stand ready to contribute donations. This is your only warning. You run, you die. I await your compliance."

A bit better then bog standard text messages like "big haul blah blah" or "Die!" during combat, no? With ED's crack audio team, I'm sure they could pull this off, there's great variety of messages for the station crew so hopefully the same would be true for NPCs.

Indeed
 
Because NPC communications are now recorded to the Journal e.g...

Code:
{ "timestamp":"2017-01-16T19:28:23Z", "event":"ReceiveText", "From":"$npc_name_decorate:#name=Chris Armstrong;",
"From_Localised":"Chris Armstrong", "Message":"$BountyHunter_HunterHostileSC_Relevant03;",
"Message_Localised":"There you are! That bounty makes you today's meal ticket.", "Channel":"npc" }


...it should be possible to write a program to parse the file in real time, extract the actual dialogue and send it to a text-to-speech engine. If you wanted to be clever you could compare each NPC name with a database of gender-weighted names and allocate a male or female voice accordingly. If you wanted to be really smart you could use a hash of the NPC's name as a seed for picking from a selection of voices, thus ensuring that the "same" NPC had the same voice each time you encountered them.

This is well beyond my meagre string-slicing skills but it seems like an ideal project for an enthusiastic coder to give it a shot if FD don't want to. Hell, if someone's already done it please let me know; I would love such a feature and a donation from my PayPal account stands ready... ;)

Is that file saved every (few) frame(s)? If its only saved on exit that wont work well. ;)

- - - Updated - - -

Until they can get procedurally generated conversations sorted and couple it with a decent voice synthesiser, no to voices. It gets tedious enough listening to SLF pilot and seeing the local pirates announcing their intentions.

No to voice synthesizer, that tech is fascinating but nowhere close enough.
 
No to voice synthesizer, that tech is fascinating but nowhere close enough.

I'm fully aware that decent voice synthesisers that can reasonably pass as human are years, if not decades, away. However, it is practically impossible to record enough voice lines due to both financial and technical limitations to give players enough variety such that they don't just keep hearing the same lines over and over and over again; without synthesis the voice the voice recording would become a limiting factor and prevent them from implementing a variety of conversations and lines. In a choice between a relatively small number prerecorded lines from a selection of VAs and text based procedural chatter, I'd take the procedural chatter every time.
 
I agree that voice messages could get old/repetitive very quickly. Perhaps the low-volume, distorted voice comms you get around stations would be better. If there are lots of NPCs around, you can hear voice comms even if you can't make out exactly what they're saying. It'd give the feeling of a populated universe, without the endless repetition.
 
i like the idea - as long as the're enough variety in the chatter - im thinking hundreds or thousands of one liners. and that it keeps evolving (heck have a contest where community contributes to one-liners and vote to choose 10 or 15 every few months) - not like the same SC music we've been hearing since the Alpha days.

This before we vocalize it; even if we never vocalize it, still this. We need CG-shipment-levels of NPC dialogue first. Nobody wants the Elite-version of Arrow-to-the-Knee guy on the comms every 5 seconds; the chatfeed repetition is bad enough.

I found the ambient chatter from OG Freelancer to be passable for its time, however. Despite eventually noticing individual VAs (looking at you nasal badguy), overall the chatter was enough to add to the feel of the game without becoming tedious. You had feed directed at your ship many times from other ships and stations, you could request someone's flight status - "This is Black 17. I'm hauling Water to the California system" - but some of it was just ships in-system chirping on open channels.
 
This before we vocalize it; even if we never vocalize it, still this. We need CG-shipment-levels of NPC dialogue first. Nobody wants the Elite-version of Arrow-to-the-Knee guy on the comms every 5 seconds; the chatfeed repetition is bad enough.

I found the ambient chatter from OG Freelancer to be passable for its time, however. Despite eventually noticing individual VAs (looking at you nasal badguy), overall the chatter was enough to add to the feel of the game without becoming tedious. You had feed directed at your ship many times from other ships and stations, you could request someone's flight status - "This is Black 17. I'm hauling Water to the California system" - but some of it was just ships in-system chirping on open channels.

I, too, think that Freelancer can serve as a baseline for quality NPC voicing. Of course, the challenge would be to take a base set of voice actors, have them read enough lines that we players won't notice too much repetition, and then run the lines through an RNG-based voice modifier to sound unique.

That way, we avoid the "Skyrim" effect.
 
So waaaaaay back when FD were laying out their "roadmap" (DDF) , voice acting was included in it. I actually think it was for one of the patches in season 3. It read something like "voice acting becomes more integral"..If I remember correctly.
 
So waaaaaay back when FD were laying out their "roadmap" (DDF) , voice acting was included in it. I actually think it was for one of the patches in season 3. It read something like "voice acting becomes more integral"..If I remember correctly.

They said voice acting is becoming important in season 3.
That's why i am guessing spaceleg is in season 3
 
Last edited:
Back
Top Bottom