Audio
The AudioComponents integrate the WebAudio-API into FUDGE. WebAudio allows for a wide range of audio application and manipulation including sound generation, effects and spatial location and emission. It is a system using so called Audio-Graphs and is completely separated from the DOM and WebGL. To learn more about the API, see https://developer.mozilla.org/en-US/docs/Web/API/Web_Audio_API
Using the Audioclasses of FUDGE, it is possible to connect the Audio-Graphs to FUDGE-Graphs to create auditive experiences matching the visual impressions.
Extends the standard AudioContext for integration with FUDGE-graphs and creates a default object at startup to be addressed as AudioManager.default. It contains a master gain, and accepts a node as the root of the graph to listen to and a ComponentAudioListener to describe the reception of the audio.
Extends the audio buffer with methods to load audio ressources (mp3).
Synchronizes audio with the FUGDE-Node it's attached to in respect to its world position and allows for specific adjustment of the volume. Effects may be added by inserting the standard Audio-Nodes of the WebAudio-API.
Serves to set the spatial location and orientation of the reception of the audio relative to the world transformation of the FUDGE-node it is attached to.
Consider a scene with the players avatar having a conversation with an NPC in a threedimensional environment. A simplified diagram explaining the connection between the FUDGE-graph with its nodes and components and the AudioContext controlled by it may look like this.
The players avatar has a ComponentAudioListener attached, since the player should hear what the avatar hears. It also contains a ComponentAudio to play the question to ask the NPC. This component controls a source-node in the AudioContext to play the audiodata, and a panner-node to locate the audio in the world. Since its in the same spot as the listener, its gain may be reduced a bit using the gain node. The NPC uses the audio-resource answer in the same manner, here probably with full gain, since it's located further away from the avatar and needs to yell. The audiocontext mixes both audiosignals, applies a master gain and sends the result to the speakers.