A library for working with musical temperaments.

Usage no npm install needed!

<script type="module">
  import temperament from 'https://cdn.skypack.dev/temperament';



npm GitHub Workflow Status

This is a library for working with musical temperaments in JavaScript. It was originally extracted from (and is used in) Temperatune, one of my other projects. I figured it could be useful on its own to other developers, so I made it into its own module.


You can install the library using NPM or similar tools (such as Yarn) which use the NPM package registry:

$ npm install temperament

You can also use a CDN, such as esm.sh, directly from a browser or Deno:

import { Temperament } from "https://esm.sh/temperament@4";

This package is ESM-only, which means it cannot be used with require.


The Temperament class encapsulates information about a temperament, providing several methods for working with processed temperament data. The constructor accepts a single argument, an object in the format described in the temperament format section:

import { Temperament } from "temperament";

let equalTemperament = new Temperament(equalTemperamentData);
equalTemperament.referencePitch = 441;
console.log(equalTemperament.getPitch("A", 4)); // Prints `441`.

In the above example, equalTemperamentData could be the object corresponding to the sample given in the basic usage section.

Temperament format

The format of a temperament is specified by the JSON schema exported as schema in index.js. The schema is also published as a JSON file with each release, and can be found with the latest release on GitHub. Each item in the schema is annotated with a description key that explains its purpose.

Some samples are included under temperaments.

Basic usage

Here is an example of a temperament file describing equal temperament:

  "name": "Equal temperament",
  "referenceName": "A",
  "referencePitch": 440,
  "referenceOctave": 4,
  "octaveBaseName": "C",
  "notes": {
    "C": ["C", 0],
    "C♯": ["C", 100],
    "D": ["C♯", 100],
    "E♭": ["D", 100],
    "E": ["E♭", 100],
    "F": ["E", 100],
    "F♯": ["F", 100],
    "G": ["F♯", 100],
    "G♯": ["G", 100],
    "A": ["G♯", 100],
    "B♭": ["A", 100],
    "B": ["B♭", 100]

All the keys used in the above example are required. The first key (and the one whose purpose is most obvious) is the name: each temperament must have a name which identifies it for use in other applications.

The next three keys describe the reference note. The reference note is commonly the one used as a tuning pitch in an ensemble: in Western music, it is typically A-4. The referenceName key gives the name of the reference note, which must correspond to a key in the notes object, and the referenceOctave key specifies the octave number of the reference note. Finally, the referencePitch is the default value (in Hz) of the reference note's pitch. The reference pitch may be configurable in user-facing applications, but you should provide a reasonable default that corresponds to common usage. For example, the reference pitch is specified above as 440 since most modern players tune to a reference of 440 Hz.

The octaveBaseName is the name of the note that should be at the bottom of each octave, and is used for octave numbering. The value of octaveBaseName must correspond to a key in the notes object. In the example above, the octaveBaseName is C, so the highest C below the reference pitch will be labelled as C-4 and the lowest C above the reference pitch will be labelled as C-5.

Finally, the bulk of the temperament is described in the notes object. The keys in this object are note names, and the corresponding values define the each note in terms of an offset (in cents) from some other note. For example, the entry "F": ["E", 100] above means that the note labelled F is 100 cents above the note labelled E. To avoid giving redundant information, the entry "C": ["C", 0] ensures that note labelled C is defined, but does not provide any information about its pitch since that information can be derived from the other notes. In general, your temperament definition should have exactly one note that is not defined relative to another note.

Conceptually, you can imagine the note information being used to deduce the pitch of each note by starting at the reference pitch (which is already known) and working outwards. For example, if we have the pitch of the reference note A, then we can immediately deduce the pitches of B♭ (100 cents above) and G♯ (100 cents below). We can then continue this process to deduce the pitches of B and G, and so on.

A note on octaves: an octave is defined to be 1200 cents, or a pitch ratio of 2:1. The notes that you specify in the notes object are assumed to fill a single octave, meaning that the definition "C♯": ["C", 1300] is equivalent to the one given in the example above for C♯, since 1300 - 1200 is 100.


In addition to the required name key, there are several other keys which can be used to add metadata to a temperament. Currently, the supported keys are as follows:

  • description: a longer description of a temperament. It is recommended to restrict this to a single sentence, with a period at the end.
  • source: a description of the source from which the temperament data was obtained. For example, this could be a URL pointing to the Wikipedia page of your temperament or another page that describes it.


This is free software, distributed under the MIT license.