suggestion for voice acting

[zdoofop]

ok, so in almost all games, the voices are scripted to omit the name of the player, because there is no way to know what name the player would pick. But what if you didn't have to know?

Sounds impossible, right? Hear me out. Text to Speech technology has come a very long way. Even Siri, Cortana, and other such voices all use TTS technology. Here's the thing: they have to record the voices somehow. What if all your voice actors had to do was make TTS voices, then you took the player's name as input and had the appropriate voice just read it?

This has other long term benefits, such as if a voice actor cannot/will not continue. You don't have to find a new one, you just take the existing tts read the new script.

 

[Torgue_Joey]

XxxBiatchexxx / *server commit suicide*

Mike1978 / Hello Mr. Mike one thousand nine huntdred seventy eight.

l llll lllll ll / barcode anyone?

B A L R O G / Hello Bee Ai El Ar Oh Gee, we need your help

 

[Nunaden]

we should use some semi synthesised voices, and a option for the player to give him-/herself a call name, and we will see who'S stupid xD

 

[Terricon4]

Or you can go this route... if you got the budget.

 

[EvilKitten]

I think this would be great if our omniframes AI talked to you. The problem with TTS though is that it tends to lack the proper inflections and tones that is present in natural speach, so I don't think it would work for actual characters talking to you. Also with that said...some names do not deserve to be spoken aloud...

The other option could be that you choose a first name from a drop down list and then have to make a unique last name? The game can refer to you by the preselected name only. I don't think it would be too bad to record a couple hundred names?

 

[zdoofop]

I just talked to someone about this, and it turns out it's harder (and more expensive) than I originally thought. So please disregard the idea.

 

[spiralofhope]

I've had this idea in mind for some time.

The solution is for the player to have a choice as to how their name is pronounced. They have their login name, and they would use a text language to better-describe how things are spoken.

login: spiralofhope
spoken: spih'rahl'of'hohpe (a bad example)

We'd be using an existing long and short vowel, and syllable concept. The game would guess, or use something basic, or perhaps as advanced as the IPA:
https://en.wikipedia.org/wiki/International_Phonetic_Alphabet

When done properly, this vocalization can be used openly, for all players and not just the account-holder.

 

[AdmiralStryker]

I was thinking give us callsign numbers for individuality. For Firefall, everyone was ARES 35. Why not let us choose a REAPER number within a range? That's a lot easier.

 

[Mahdi]

Ahhh Stryker, I think you are on to something. This could add an element to the flare/s.o.s. aspect that was talked about recently.

 

[Nalessa]

Problem is, text to speech still has a somewhat lacking emotional tone to it.

Reason I liked Aero for example was because she felt like a human being and has reaction that are fitting, I'd rather have that as my advisor/adjutant/whatever, then an emotionless AI.

 

[0V3RKILL]

I wouldn't mind having the same actors from firefall. they did a very good job.

 

[Nalessa]

I wouldn't mind having the same actors from firefall. they did a very good job.

Except cupcake, no more cupcake pls :<

 

[AdmiralStryker]

Problem is, text to speech still has a somewhat lacking emotional tone to it.

Reason I liked Aero for example was because she felt like a human being and has reaction that are fitting, I'd rather have that as my advisor/adjutant/whatever, then an emotionless AI.

Give us a voiced selection to choose from. That massively cuts the workload, even if it's 4 options.

 

[Grummz]

I have someone I can talk to about this. He helped write Microsoft's Cortana. I'll ask about how hard it is.

 

[Sik San]

Is that rly a thing for many people? Not that necessary, imo.

 

[AdmiralStryker]

Is that rly a thing for many people? Not that necessary, imo.

It's nice for immersion. Not 100% vital, but a nice quality of life improvement.

 

[Nalessa]

I have someone I can talk to about this. He helped write Microsoft's Cortana. I'll ask about how hard it is.

Is Laura Post going to be voice acting something?

Just wondering, sucks how she eventually got ignored to do Aero, would be cool to give her a new opportunity with this game!

 

[AdmiralStryker]

Is Laura Post going to be voice acting something?

Just wondering, sucks how she eventually got ignored to do Aero, would be cool to give her a new opportunity with this game!

All M3 backers are getting a voice pack in game done by her.

 

[EvilKitten]

All M3 backers are getting a voice pack in game done by her.

While awesome in and of itself I am hoping she will have a much larger roll than ~2000 players. Perhaps the AI can be based on a real in game character who is also part of the story/action. I remember when Aero appeared "in person" in the ...blackwater(?) mission back in beta and I thought it was the coolest thing ever to suddenly run face to face with the person who always "talked in my head".

 

[MasterGeneralB]

I've had this idea in mind for some time.

The solution is for the player to have a choice as to how their name is pronounced. They have their login name, and they would use a text language to better-describe how things are spoken.

login: spiralofhope
spoken: spih'rahl'of'hohpe (a bad example)

We'd be using an existing long and short vowel, and syllable concept. The game would guess, or use something basic, or perhaps as advanced as the IPA:
https://en.wikipedia.org/wiki/International_Phonetic_Alphabet

When done properly, this vocalization can be used openly, for all players and not just the account-holder.

That was literally what I was thinking of. The only proper way to get ANY name right, no matter how horribly bad it is (mashup of an entire sentence anyone?) is to provide its phonetic pronunciation. Allow for a playback to be performed by the player until they are satisfied with the result. This can be done quite easily and with relatively low overhead as it would all be done client-side. Team-speak is a good example of a program that has this feature, except it is even more rudimentary in that it doesn't require syllable parsing. Suffice to say its relatively easy to work with. It's not that difficult to have a recorded dialog/speech in-game that provides a blank gap for any name to be inserted... with the inserted name being TTS'ed, if not attempting to just TTS all the dialog.