I’m making a multiplayer space shooter - and voice chat seems pretty important to me. Anyone have experience with both of these assets? Which would you say is better?
Sound quality isn’t as important as ease of implementation. I’d like positional sound (So that the audio seems to be coming from the individual player ships, possibly with doppler effect).
I’m using default Unity networking.
Opinions? I see that uSpeak is $10 cheaper. Might be all it takes to lean me into purchase - but I want to hear some opinions first!
If you want ease of implementation then the person that wrote the more expensive package has given alot more free advice on these forums so is likely to help you more that the writer of the other package. If someone has actual experience with both packages feel free to chime in.
Also it seems alls these special effects you are want to transmit over VoiP are not possible. It’s an audio stream not a logic stream. It’s just a thin-band sound stream and both the sounds and the sound effects are better left in the game and handled by game / Unity networking logic. You want your game and it’s assets to be self contained and you want to design it so that changes in game state are minimized and transmitted via Unity (or other networking packages). You do not want to transmit FX and music sounds, meshes, textures, and such. Designing this way will also lend itself to saving games for later resumption. Please see Mr Talbot’s Unity Serializer / Game Save asset and learn to use that and then see that what you want to transmit in a networked game will be similar to that technique but as a series of deltas.
Use VoiP for friends to talk telephone or PTT style and game / network logic to position the incoming voice in your game with Unity’s audio effects system.
OK, take what I say with a grain of salt (I am probably biased, as the creator of USpeak, just a fair warning)
I think USpeak is probably easier to set up, there is already a built-in example with Unity Networking. Fholm’s pack involves some setup with proxy server game objects or something (I don’t quite understand it, but if I were to sit down and actually get it running it would probably be more clear). But with USpeak you pretty much just have to send the data through RPC and then feed it back to USpeak with only a few lines of code.
As far as quality, I believe fholm’s pack is definitely better / more clear (his includes Speex, and mine does too but I think he’s got a better implementation than I do). On the other hand, if quality isn’t an issue so long as people can understand each other my package has a default codec of ADPCM which has just about the same bandwidth, and is a lot more CPU friendly (but much lower quality and scratchier, not nearly as bad as Call of Duty however). And if bandwidth isn’t an issue there’s also Mu-Law which sounds way better than ADPCM at about twice the bandwidth.
