The Music Maker Project (MMP)

Introduction

The Music Maker Project (MMP) is a Domain-Specific Language (DSL) that allows you to quickly create basic sound effects and export them as WAV files.

Objective

The objective of this DSL is to enable software developers that have negligible/very basic knowledge of music to create basic sound effects for use in small-scale software projects (such as a simple quiz game).

Technical Context

MMP is implemented using a modular, event-based approach and relies on the JavaScript's native Web Audio API standard for producing and recording sounds. This is an immensively powerful, ever-evolving standard and provides plently of opportunity for future extensibility/devlopment of this project.

The Basics

Structure

The DSL consists of 3 major blocks:

• Sound
• Combination
• Play

Sound

The sound block describes individual audio tracks to be used for creating your sound effect.

There are 2 types of sound blocks:

• Preset
• Clip

Preset

Preset sound blocks represent the use of an exisiting sound effect.

Sound Kick1 as Preset
    Pattern: x-xx-x
    Repeat: 1
    UseSound: "Kick"
EndSound

You specify which sound effect you want to use in the UseSound attribute. Currently, you can choose from Kick, Bass, OpenHats, ClosedHats & Snare.

Clip

Clip sound blocks represent the use of a sequence of musical notes.

Sound MyMelody as Clip
    Pattern: x
    Repeat: 2
    UseSound: "C1 D2 F7"
EndSound

You specify the sequence of notes (with octaves) in the UseSound attribute as shown in the example sound block, named 'MyMelody', above.

In both types of sound blocks, you must also specify Pattern & Repeat.

Pattern is denoted by a series of 'x' and '-'. x means that the sound specified in UseSound should be played and '-' represents a pause of (approximately) 0.5 seconds.

Repeat denotes how many times you wish to repeat the pattern.

Combination

This block specifies how you would like to sequence the sounds you defined earlier.

Combination Draft1
    Track
        Volume: 75
        Components: [5*Kick1, 1*MyMelody]
    EndTrack
    Track
        Offset: 100
        Volume: 25
        Components: [2*MyMelody]
    EndTrack
EndCombination

The two main parts of a Combination are the Track and Components.

As is illustrated by the way the DSL is strucuted, all Tracks in a combination are played in parallel.

In a Track, you specify:

• Volume - loudness of the track.
• Offset - start time delay in milliseconds

and

• Components

Components represents the sounds that a track should play. The sounds specified in the components section are played in sequence (one after the other) for each track.

Play

To help you ideate/draft your sound effect, you are allowed to create multiple combination blocks in your code. However, only one of them will played/rendered upon executuion.

The Play block is used to specify which combination block you would like to be played in the current execution.

Play Draft1

Full Example

Sound Kick1 as Preset
    Pattern: x-xx-x
    Repeat: 1
    UseSound: "Kick"
EndSound

Sound Melody1 as Clip
    Pattern: x
    Repeat: 2
    UseSound: "C1 D2 F7"
    EndSound

Combination Draft1
    Track
        Volume: 75
        Components: [5*Kick1, 1*Melody1]
    EndTrack
    Track
        Offset: 100
        Volume: 25
        Components: [2*Melody1]
    EndTrack
EndCombination
Play Draft1

Documenation

program         : sound+ combination+ play
sound           : 'Sound ' soundname 'as ' SUBTYPE pattern repeat usesound 'EndSound'
soundname       : TEXT
pattern         : 'Pattern:' TEXT
repeat          : 'Repeat:' NUM
usesound        : 'UseSound:' USESOUNDS '"'
combination     : 'Components:' TEXT track+ 'EndCombination'
track           : 'Track' offset? volume? components* 'EndTrack'
offset          : 'Offset:' NUM
volume          : 'Volume:' NUM
components      : 'Components:' '[' component [','component]* ']'
component       : NUM? '*' COMPONENT_NAME
play            : 'Play' TEXT

TEXT            : ~[[\]\r\n ]+
SUBTYPE         : 'Preset' | 'Clip';
NUM             : [0-9]+
USESOUNDS       : ~[[\]'"]+
COMPONENT_NAME  : ~[[\]\r\n,* ]+

In addition to the grammar rules above, a few logical rules are in place in order for the DSL to run properly:

• The pattern must be a string consisting of any combination of only characters "x" and "-" and no other characters, where an "x" means to play a note, where as "-" means rest (don't play a note)
• TEXT in each component must match a soundname created by you
• If a sound has SUBTYPE "Clip", USESOUNDS must be a single line of text, with each note following the pattern of {note}{octave} with a space between each note(e.g. "C2 G9")
• If a sound has SUBTYPE "Preset", then the USESOUNDS must match exactly with any one of the following sound names provided: "Kick", "Base", "Snare", "OpenHats", or "ClosedHats"
• The indentation level on each line does not affect the success/failure of running the DSL.