Slakh.md

Slakh Data Card

Dataset Overview

Size of dataset	Number of audios	Duration
81 GB	18276	1 374 hrs

Data Collection

Source	Collecting Method
Slakh2100 Zenodo page	in Slakh Zenodo page, download slakh2100_flac_redux.tar.gz. You can find the original website here

Preprocessing Principles

You may refer to preprocess_slakh.py for all the details. Here we just offer a concise summary:

Overview

Some audio-json pairs selected from the processed dataset below:

{
    "text": "playing piano music synthesized with scarbee clavinet full plugin",
    "tag": [
        "piano",
        "electric piano 1"
    ],
    "original_data": {
        "audio_rendered": true,
        "inst_class": "Piano",
        "integrated_loudness": -21.746239958311875,
        "is_drum": false,
        "midi_program_name": "Electric Piano 1",
        "midi_saved": true,
        "plugin_name": "scarbee_clavinet_full.nkm",
        "program_num": 4,
        "filename": "Track01595/stems/S04.flac"
    }
}

I. Json file generation principles

• text entry If a mixfile was used, text used was playing mix of {instrument_name1} {instrument_name2} ... and stems use playing {instrument_name} music synthesized with {plugin_name} plugin.
• tag entry We use instrument name and MIDI program name as tags.
• original data We save filename, instrument class, MIDI program name, plugin name for every audio as well as audio duration, the dataset name and dataset description.

II. Audio filtering principles

Discard all audios failed to be read by soundfile.read() method or denied by FFmpeg while processing.

III. Audio format specifications

After the preprocessing work, all audio files should be in FLAC format with sampling rate of 48KHZ. (Processed by ffmpeg).