Maqsudjonpolatov commited on
Commit
f1d5966
·
verified ·
1 Parent(s): 1b88521

Dataset card: 20 synthetic tracks with stems, MIDI and exact annotations

Browse files
Files changed (1) hide show
  1. README.md +156 -0
README.md CHANGED
@@ -1,3 +1,159 @@
1
  ---
2
  license: cc-by-4.0
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
3
  ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  ---
2
  license: cc-by-4.0
3
+ pretty_name: Nullsample Multitrack
4
+ task_categories:
5
+ - audio-classification
6
+ - audio-to-audio
7
+ tags:
8
+ - music
9
+ - synthetic
10
+ - procedural-audio
11
+ - multitrack
12
+ - stems
13
+ - source-separation
14
+ - music-transcription
15
+ - drum-transcription
16
+ - beat-tracking
17
+ - chord-recognition
18
+ - key-detection
19
+ - midi
20
+ - mir
21
+ size_categories:
22
+ - n<1K
23
+ configs:
24
+ - config_name: default
25
+ data_files:
26
+ - split: train
27
+ path: "mix/*"
28
  ---
29
+
30
+ # Nullsample Multitrack
31
+
32
+ **20 complete tracks generated from code, with zero samples. Each comes with its mix, six
33
+ stems, a MIDI file and exact annotations**: tempo, downbeats, key, the chord in every bar, the sections,
34
+ every drum hit by instrument and every note of every part. That is 40,060 drum hits and 32,919
35
+ notes over 37 minutes of music.
36
+
37
+ The labels are not estimated and not hand-annotated. They are the score the audio was synthesised from,
38
+ so they are exact by construction. No recordings, samples or trained models were used. Every sound is
39
+ computed from oscillators, noise, envelopes and filters by
40
+ [Nullsample](https://nullsample.maqsudjon.com) ([source](https://github.com/maqsudjon-cell/nullsample),
41
+ AGPL-3.0). Nothing in the audio came from anyone else's recording.
42
+
43
+ **Every track can be reproduced bit for bit from its seed**, on any machine. The engine avoids the
44
+ platform-dependent `Math.sin`/`exp`/`pow` and is tested byte-identical across Node and browsers. You can
45
+ listen to any seed at `https://nullsample.maqsudjon.com/generate/#s=<seed>`, or render the lossless WAV
46
+ yourself:
47
+
48
+ ```bash
49
+ git clone https://github.com/maqsudjon-cell/nullsample && cd nullsample && git checkout b6adf10
50
+ npm install && npm run render -- --seed NULL-0001 --out NULL-0001.wav
51
+ npm run dataset -- --out ./dataset # this whole dataset
52
+ ```
53
+
54
+ The audio came from engine commit `c912de6` (`engine_commit` in every annotation file). The
55
+ checkout above has the same engine plus the script that wrote this dataset.
56
+
57
+ ## What it is good for
58
+
59
+ - **Source separation:** six stems per track (drums, 808 bass, lead, arp, pads, fx).
60
+ - **Drum and note transcription:** onsets and notes that are exact to the sample.
61
+ - **Beat, downbeat and tempo tracking.**
62
+ - **Key and chord recognition, and structure segmentation.**
63
+ - **Testing an MIR tool against ground truth** before trusting it on real recordings.
64
+
65
+ ## Files
66
+
67
+ | Path | What |
68
+ |---|---|
69
+ | `mix/<seed>.mp3` | the track, MP3 192 kbps, 44.1 kHz stereo |
70
+ | `mix/metadata.csv` | one row per track: tempo, key, scale, bars, length, form, sections, chords. The viewer reads it. |
71
+ | `stems/{drums,bass808,lead,arp,pads,fx}/<seed>.mp3` | each part rendered on its own, MP3 128 kbps |
72
+ | `annotations/<seed>.json` | the labels, times in seconds |
73
+ | `midi/<seed>.mid` | the same notes as a Standard MIDI File |
74
+
75
+ ### `annotations/<seed>.json`
76
+
77
+ ```json
78
+ {
79
+ "seed": "CLIP-N0ISE", "tempo_bpm": 152.5, "time_signature": "4/4", "bars": 68, "duration": 109.516,
80
+ "downbeats": [0, 1.5738, 3.1475, ...],
81
+ "key": { "tonic": "G", "scale": "phrygian", "scale_intervals": [0,1,3,5,7,8,10] },
82
+ "form": "double-drop",
83
+ "sections": [{ "name": "intro", "start": 0, "end": 6.2951, "intensity": 0.12 }, ...],
84
+ "chords": [{ "start": 0, "end": 1.5738, "chord": "Gsus2", "root": "G", "voicing": "sus2", "notes": [55,57,62] }, ...],
85
+ "drums": { "kick": [{ "time": 6.2951, "velocity": 1 }, ...],
86
+ "hatClosed": [{ "time": 6.2951, "velocity": 0.825 }, ...],
87
+ "hatOpen": [{ "time": 6.4918, "velocity": 0.629 }, ...],
88
+ "rim": [{ "time": 6.4918, "velocity": 0.317 }, ...],
89
+ "snare": [{ "time": 6.6885, "velocity": 0.995 }, ...],
90
+ "clap": [{ "time": 7.4754, "velocity": 1 }, ...] },
91
+ "notes": { "bass808": [{ "start": 5.9016, "end": 6, "midi": 24, "velocity": 0.686 }, ...],
92
+ "lead": [{ "start": 6.2951, "end": 6.3934, "midi": 94, "velocity": 1 }, ...],
93
+ "arp": [{ "start": 37.7703, "end": 37.8146, "midi": 110, "velocity": 0.95 }, ...],
94
+ "pads": [{ "start": 0, "end": 1.3849, "midi": 67, "velocity": 0.8 }, ...] },
95
+ "fx": [{ "kind": "riser", "start": 3.1475, "end": 6.2951 }, ...],
96
+ "regenerate": "https://nullsample.maqsudjon.com/generate/#s=CLIP-N0ISE",
97
+ "engine_commit": "c912de6da8dc5c66871a8acb8c0441a2d4666193"
98
+ }
99
+ ```
100
+
101
+ The MIDI files have 480 ticks per beat and the track's tempo. They have a tempo track with the sections
102
+ as markers, the drums on channel 10 (General MIDI: kick 36, rim 37, snare 38, clap 39, closed hat 42,
103
+ open hat 46), and one track each for `bass808`, `lead`, `arp` and `pads`. Velocities in the JSON are
104
+ 0–1. The pads are chords with no velocity of their own, written at 0.8. Chord names: `5` is a power chord, `m` minor, `m7`, `sus2`, `sus4`, `add9`, `m(add9)`.
105
+
106
+ ## The tracks
107
+
108
+ Tempo 151–159.5 BPM. Scales: phrygian, minorPentatonic, naturalMinor, harmonicMinor, phrygianDominant.
109
+
110
+ | Seed | BPM | Key | Form | Length | Sections |
111
+ |---|---|---|---|---|---|
112
+ | `CLIP-N0ISE` | 152.5 | G phrygian | double-drop | 1:50 | intro drop break drop2 break2 drop3 outro |
113
+ | `DUCK-UNDER` | 155 | F minorPentatonic | double-drop | 1:48 | intro drop break drop2 break2 drop3 outro |
114
+ | `FOLD-BACK3` | 159 | C# minorPentatonic | double-drop | 1:45 | intro drop break drop2 break2 drop3 outro |
115
+ | `GLASS-EDGE` | 156.5 | G naturalMinor | cold-open | 1:47 | drop verse drop2 verse2 drop3 outro |
116
+ | `GRID-LOCK4` | 159.5 | A# harmonicMinor | long-build | 1:57 | intro build drop break drop2 outro |
117
+ | `HARD-CODE1` | 158 | A naturalMinor | double-drop | 1:46 | intro drop break drop2 break2 drop3 outro |
118
+ | `K7M2-QX4B` | 154.5 | D harmonicMinor | long-break | 1:42 | intro build drop break drop2 outro |
119
+ | `LAST-CALL9` | 154 | B naturalMinor | cold-open | 1:48 | drop verse drop2 verse2 drop3 outro |
120
+ | `NULL-0001` | 155 | E minorPentatonic | sustained | 1:54 | intro hold lift hold2 lift2 outro |
121
+ | `OSC-ILLATE` | 156.5 | A harmonicMinor | cold-open | 1:47 | drop verse drop2 verse2 drop3 outro |
122
+ | `PHRY-GIAN9` | 156 | E phrygian | straight-in | 1:53 | drop break drop2 bridge drop3 outro |
123
+ | `PRIME-DLAY` | 155.5 | D harmonicMinor | arc | 1:54 | intro build drop break drop2 bridge outro |
124
+ | `RAGE-88KK` | 156.5 | C# harmonicMinor | sustained | 1:53 | intro hold lift hold2 lift2 outro |
125
+ | `SAW-STACK5` | 155 | E naturalMinor | cold-open | 1:48 | drop verse drop2 verse2 drop3 outro |
126
+ | `SEED-4242X` | 154 | D# phrygian | straight-in | 1:55 | drop break drop2 bridge drop3 outro |
127
+ | `SINE-WAVE7` | 159.5 | F# naturalMinor | sustained | 1:51 | intro hold lift hold2 lift2 outro |
128
+ | `SUB-DROP01` | 158 | A harmonicMinor | long-build | 1:58 | intro build drop break drop2 outro |
129
+ | `TAPE-NULL2` | 151 | F phrygianDominant | long-break | 1:44 | intro build drop break drop2 outro |
130
+ | `VOID-7X2A` | 151.5 | C naturalMinor | long-break | 1:44 | intro build drop break drop2 outro |
131
+ | `ZERO-SAMPL` | 159 | A phrygian | straight-in | 1:51 | drop break drop2 bridge drop3 outro |
132
+
133
+ ## Limits
134
+
135
+ - **One style.** All tracks come from one preset (energetic electronic, hyperpop-like), so this is
136
+ not a sample of music in general. It is small: use it to test and probe models, or as a clean
137
+ supplement, not as a training corpus on its own.
138
+ - **Stems do not sum exactly to the mix.** Each stem is its bus rendered alone through the master
139
+ chain at a fixed drive. The mix also has mix-wide effects (tape stops, stutters) and a limiter
140
+ that reacts to everything at once.
141
+ - **MP3, to keep it small.** Annotation times come from the engine's sample positions. Decoders
142
+ that honour the LAME gapless header, such as ffmpeg and most libraries, line up within a few
143
+ milliseconds. For lossless audio, render the seed.
144
+ - The synthesis parameters behind each sound are not included.
145
+
146
+ ## Licence
147
+
148
+ The audio, MIDI and annotations: **CC BY 4.0**. Please credit "Nullsample (nullsample.maqsudjon.com)".
149
+ The engine that made them: [AGPL-3.0](https://github.com/maqsudjon-cell/nullsample). Audio you render
150
+ yourself with it is yours.
151
+
152
+ ```bibtex
153
+ @misc{nullsample_multitrack,
154
+ title = {Nullsample Multitrack: synthetic music with exact annotations},
155
+ author = {Polatov, Maqsudjon},
156
+ year = {2026},
157
+ url = {https://nullsample.maqsudjon.com}
158
+ }
159
+ ```