Dataset Viewer
Auto-converted to Parquet Duplicate
audio
audioduration (s)
39.4
462
filename
stringlengths
19
19
track_id
stringlengths
10
10
Track00049_mix.flac
Track00049
Track00077_mix.flac
Track00077
Track00146_mix.flac
Track00146
Track00170_mix.flac
Track00170
Track00212_mix.flac
Track00212
Track00232_mix.flac
Track00232
Track00240_mix.flac
Track00240
Track00251_mix.flac
Track00251
Track00256_mix.flac
Track00256
Track00280_mix.flac
Track00280
Track00296_mix.flac
Track00296
Track00299_mix.flac
Track00299
Track00350_mix.flac
Track00350
Track00356_mix.flac
Track00356
Track00365_mix.flac
Track00365
Track00366_mix.flac
Track00366
Track00386_mix.flac
Track00386
Track00401_mix.flac
Track00401
Track00422_mix.flac
Track00422
Track00424_mix.flac
Track00424
Track00425_mix.flac
Track00425
Track00450_mix.flac
Track00450
Track00470_mix.flac
Track00470
Track00476_mix.flac
Track00476
Track00486_mix.flac
Track00486
Track00495_mix.flac
Track00495
Track00504_mix.flac
Track00504
Track00521_mix.flac
Track00521
Track00536_mix.flac
Track00536
Track00539_mix.flac
Track00539
Track00554_mix.flac
Track00554
Track00566_mix.flac
Track00566
Track00602_mix.flac
Track00602
Track00611_mix.flac
Track00611
Track00613_mix.flac
Track00613
Track00620_mix.flac
Track00620
Track00623_mix.flac
Track00623
Track00640_mix.flac
Track00640
Track00642_mix.flac
Track00642
Track00646_mix.flac
Track00646
Track00647_mix.flac
Track00647
Track00675_mix.flac
Track00675
Track00678_mix.flac
Track00678
Track00688_mix.flac
Track00688
Track00697_mix.flac
Track00697
Track00699_mix.flac
Track00699
Track00706_mix.flac
Track00706
Track00714_mix.flac
Track00714
Track00715_mix.flac
Track00715
Track00719_mix.flac
Track00719
Track00720_mix.flac
Track00720
Track00732_mix.flac
Track00732
Track00744_mix.flac
Track00744
Track00750_mix.flac
Track00750
Track00757_mix.flac
Track00757
Track00762_mix.flac
Track00762
Track00764_mix.flac
Track00764
Track00773_mix.flac
Track00773
Track00790_mix.flac
Track00790
Track00796_mix.flac
Track00796
Track00800_mix.flac
Track00800
Track00810_mix.flac
Track00810
Track00820_mix.flac
Track00820
Track00821_mix.flac
Track00821
Track00829_mix.flac
Track00829
Track00833_mix.flac
Track00833
Track00847_mix.flac
Track00847
Track00848_mix.flac
Track00848
Track00851_mix.flac
Track00851
Track00862_mix.flac
Track00862
Track00877_mix.flac
Track00877
Track00884_mix.flac
Track00884
Track00895_mix.flac
Track00895
Track00899_mix.flac
Track00899
Track00916_mix.flac
Track00916
Track00919_mix.flac
Track00919
Track00925_mix.flac
Track00925
Track00933_mix.flac
Track00933
Track00934_mix.flac
Track00934
Track00935_mix.flac
Track00935
Track00942_mix.flac
Track00942
Track00956_mix.flac
Track00956
Track00960_mix.flac
Track00960
Track00961_mix.flac
Track00961
Track00962_mix.flac
Track00962
Track00964_mix.flac
Track00964
Track00969_mix.flac
Track00969
Track00981_mix.flac
Track00981
Track00982_mix.flac
Track00982
Track00990_mix.flac
Track00990
Track00997_mix.flac
Track00997
Track01000_mix.flac
Track01000
Track01011_mix.flac
Track01011
Track01020_mix.flac
Track01020
Track01028_mix.flac
Track01028
Track01046_mix.flac
Track01046
Track01052_mix.flac
Track01052
Track01061_mix.flac
Track01061
Track01066_mix.flac
Track01066
Track01067_mix.flac
Track01067
End of preview. Expand in Data Studio

Slakh2100

Quick Start

from huggingface_hub import snapshot_download

# Download the full dataset (~500GB extracted)
path = snapshot_download(
    repo_id="schismaudio/slakh2100",
    repo_type="dataset",
)

Dataset Description

Slakh2100 (Synthesized Lakh) is a large-scale multi-track music dataset containing 2,100 automatically mixed tracks with isolated instrument stems and aligned MIDI. Created by Manilow et al. (2019) at Northwestern University, Slakh uses MIDI files from the Lakh MIDI Dataset rendered through professional-grade VST instruments to produce realistic multi-track audio.

Each track contains isolated stems for every instrument (drums, bass, guitar, piano, strings, etc.), making Slakh the standard benchmark dataset for music source separation research. The "redux" version hosted here uses lossless FLAC compression to reduce the download size from the original ~500GB WAV to ~105GB.

Dataset Structure

Directory Layout

slakh2100_flac_redux/
  Track00001/
    mix.flac                  # Full mix of all instruments
    stems/
      S00.flac               # Stem 0 (e.g., drums)
      S01.flac               # Stem 1 (e.g., bass)
      S02.flac               # Stem 2 (e.g., piano)
      ...
    MIDI/
      S00.mid                # MIDI for stem 0
      S01.mid                # MIDI for stem 1
      ...
    metadata.yaml            # Instrument labels, plugin info, mix parameters
  Track00002/
    ...

Data Fields

Field Type Description
mix.flac Audio Full mix of all instrument stems (44.1kHz stereo FLAC)
stems/S*.flac Audio Isolated instrument stems (44.1kHz stereo FLAC)
MIDI/S*.mid MIDI Aligned MIDI file for each stem
metadata.yaml YAML Instrument labels, VST plugin names, mixing parameters

Data Splits

Split Tracks Description
train 1,500 Training set
validation 375 Validation set
test 225 Test set

Splits are defined by directory structure. Each split contains non-overlapping tracks.

Audio Format

  • Sample rate: 44.1 kHz
  • Channels: Stereo
  • Format: FLAC (lossless compression)
  • Average stems per track: ~5-10 (varies by arrangement complexity)

Instrument Taxonomy

Slakh maps 128 General MIDI instrument programs to 34 instrument classes. Common classes include:

Class Instruments
Drums All MIDI percussion (channel 10)
Bass Acoustic Bass, Electric Bass, Synth Bass
Piano Acoustic Piano, Electric Piano
Guitar Acoustic Guitar, Electric Guitar (clean/distorted)
Strings Violin, Viola, Cello, String Ensemble
Brass Trumpet, Trombone, French Horn
Reed Saxophone, Clarinet, Oboe
Synth Lead Various synthesizer lead patches
Synth Pad Various synthesizer pad patches

The full mapping is defined in each track's metadata.yaml.

Usage Examples

Load a track and its stems

import soundfile as sf
import yaml
from pathlib import Path

track_dir = Path("slakh2100_flac_redux/Track00001")

# Load the full mix
mix, sr = sf.read(track_dir / "mix.flac")

# Load metadata to identify instruments
with open(track_dir / "metadata.yaml") as f:
    meta = yaml.safe_load(f)

# Load individual stems
for stem_name, stem_info in meta["stems"].items():
    audio, sr = sf.read(track_dir / "stems" / f"{stem_name}.flac")
    print(f"{stem_name}: {stem_info['inst_class']} ({audio.shape[0] / sr:.1f}s)")

Extract drum stems for transcription

import yaml
import soundfile as sf
from pathlib import Path

slakh_root = Path("slakh2100_flac_redux")

for track_dir in sorted(slakh_root.glob("Track*")):
    with open(track_dir / "metadata.yaml") as f:
        meta = yaml.safe_load(f)
    for stem_name, stem_info in meta["stems"].items():
        if stem_info["inst_class"] == "Drums":
            audio, sr = sf.read(track_dir / "stems" / f"{stem_name}.flac")
            midi_path = track_dir / "MIDI" / f"{stem_name}.mid"
            print(f"{track_dir.name}/{stem_name}: drums ({audio.shape[0] / sr:.1f}s)")

Dataset Creation

Source Data

Slakh is built from the Lakh MIDI Dataset (LMD), a collection of 176,581 unique MIDI files. The 2,100 tracks in Slakh were selected from LMD to maximize instrument diversity and arrangement complexity. Each MIDI file was rendered through professional-grade VST instruments using a custom synthesis pipeline.

Synthesis Pipeline

MIDI tracks were assigned to specific VST instruments based on their General MIDI program numbers. The rendering pipeline uses high-quality commercial and open-source VST plugins to produce audio that sounds significantly more realistic than General MIDI synthesis. Each instrument is rendered as an isolated stem, then mixed together to create the full track.

Annotations

MIDI files serve as ground-truth annotations. Since the audio is synthesized directly from MIDI, the alignment between audio and MIDI is exact (sample-accurate). No manual annotation was needed.

Known Limitations

  • Synthesized audio only: Despite using professional VST instruments, the audio is synthesized from MIDI. Real recordings have room acoustics, bleed between instruments, and performance nuances not captured by MIDI.
  • MIDI-derived performances: Dynamics and timing are limited to what MIDI captures. Subtle expressive details (ghost notes, brush work, muting techniques) may be missing or simplified.
  • Pop/rock bias: The Lakh MIDI Dataset skews toward popular Western music genres. Underrepresented genres include jazz, classical, and non-Western music.
  • Variable stem quality: Some VST renderings are more realistic than others. Certain instrument classes (e.g., acoustic guitar, vocals) are particularly difficult to synthesize convincingly.
  • No vocals: The dataset does not contain vocal stems, as MIDI cannot represent singing.

Related Datasets

This dataset is part of the Drum Audio Datasets collection by schismaudio. Related datasets:

Citation

@inproceedings{manilow2019cutting,
  title={Cutting Music Source Separation Some Slakh: A Dataset to Study the Impact of Training Data on Accuracy},
  author={Manilow, Ethan and Wichern, Gordon and Seetharaman, Prem and Le Roux, Jonathan},
  booktitle={Proceedings of the International Society for Music Information Retrieval Conference (ISMIR)},
  year={2019}
}

License

This dataset is released under the Creative Commons Attribution 4.0 International License (CC-BY 4.0).

You are free to share and adapt this dataset for any purpose, including commercial use, as long as you give appropriate credit to the original authors (Manilow et al., Northwestern University).

Downloads last month
128

Collection including schismaudio/slakh2100