One of the largest openly-available, structured, de-duplicated lists of musical instruments anywhere — 8,240 entries covering 4,448 unique canonical instruments from every inhabited continent and musical tradition, classified by the Hornbostel–Sachs system, with a cited source for every row.
Built by systematic enumeration across encyclopedic and museum sources — not AI free-recall, which is why it goes far past the ~1,000–2,000 ceiling typical of hand- or model-built lists.
| Total entries | 8,240 |
| Unique canonical instruments | 4,448 |
| Chordophones (strings) | 2,185 |
| Aerophones (wind) | 2,658 |
| Idiophones (struck self-sounding) | 1,746 |
| Membranophones (drums) | 1,095 |
| Electrophones (electric/electronic) | 546 |
| File | Best for |
|---|---|
musical_instruments.csv |
Universal import — Excel, pandas, R, anything |
musical_instruments.json |
Apps & APIs (array of objects, UTF-8) |
musical_instruments.db |
Drop-in SQLite database, indexed and query-ready |
musical_instruments.xlsx |
Spreadsheet users (sorted, styled) |
musical_instruments.txt |
Plain-text reading, grep, terminals |
musical_instruments.pdf |
Printable / browsable reference (grouped by class) |
All formats are generated from the same data and kept in sync.
| Column | Meaning |
|---|---|
id |
Row number (1…8240) |
name |
The instrument's name (this row) |
canonical |
Canonical parent instrument; equals name if the row is the canonical |
variant_type |
distinct · subvariant · altname · spelling (see below) |
region |
Culture / country / tradition |
hs_class |
chordophone · aerophone · idiophone · membranophone · electrophone · unknown |
hs_number |
Hornbostel–Sachs number where known |
notes |
Short description |
source |
Where the entry was verified |
Variant types — every row maps to a canonical parent so you can collapse or expand the list:
- distinct — a distinct instrument
- subvariant — a named size / range / regional sub-type (e.g. bass mandolin → mandolin)
- altname — an alternate or regional name for the same instrument
- spelling — a transliteration / spelling variant
To get only distinct instruments, filter variant_type = 'distinct' (or use the canonical_instruments SQL view).
SQLite
-- every distinct instrument, alphabetical
SELECT name, region, hs_class FROM canonical_instruments;
-- all string instruments from China
SELECT name, notes FROM instruments
WHERE hs_class='chordophone' AND region LIKE '%China%';
-- collapse a family: every name that maps to "mandolin"
SELECT name, variant_type FROM instruments WHERE canonical='Mandolin';Python (pandas)
import pandas as pd
df = pd.read_csv("musical_instruments.csv")
distinct = df[df.variant_type == "distinct"]
print(df.groupby("hs_class").size())JavaScript / Node
const data = require("./musical_instruments.json");
const winds = data.filter(d => d.hs_class === "aerophone");A multi-round, multi-agent web-enumeration pipeline:
- Backbone — Wikipedia's List of musical instruments by Hornbostel–Sachs number leaf tables + dozens of per-culture lists.
- Round 1 (broad) — ~59 research passes by culture, taxonomic family, and instrument databases (Wikipedia, MIMO, Britannica), each de-duplicated against the growing list.
- Round 2 (long tail) — ~46 finer passes targeting ethnic-minority traditions, historical/archaeological eras, systematic family sub-variants, and electronic-by-model.
Every candidate is normalised (diacritics stripped, lower-cased) and de-duplicated against the existing set before being added, with provenance recorded in source.
Compiled primarily from Wikipedia, MIMO (Musical Instruments Museums Online), Encyclopædia Britannica, Wiktionary, and assorted ethnomusicology / organology references (per-row in the source column).
This is a large, defensible compilation — not a claim of absolute completeness. Obscure regional and historical instruments certainly exist that aren't here yet. Classifications and spellings follow the cited sources and may vary between scholars. Corrections and additions are welcome.
Open an issue or PR. The repo is regenerated from the data by the build scripts; please add rows to the source data with a source, and keep the canonical mapping consistent.
Released under Creative Commons Attribution-ShareAlike 4.0 (CC BY-SA 4.0) — consistent with its Wikipedia-derived content. You may share and adapt for any purpose, including commercially, with attribution and under the same license. See LICENSE.