Drums to MIDI · ADTOF Plus / Direct

Choose ADTOF Plus for a complete mix: it runs MDX23C drum separation, five-stem drum separation, ADTOF transcription, and velocity/hi-hat post-processing. Choose ADTOF PyTorch Direct for an already isolated drum stem.

15 credits/useFile inputMax 100 MBEstimated 20-1800 seconds

Input

Supports inputs in formats such as mp3 / wav / flac / ogg / m4a / aac / aif / aiff / mp4 / mov / mkv

Supported formats

mp3wavflacoggm4aaacaifaiffmp4movmkv

Output

MIDI

Use case

Edit notes, instrumentation and performance details.

Models, parameters, and sources

ADTOF PlusADTOF Direct · PyTorchADTOF Plus source (pinned)ADTOF PyTorch source (pinned)ADTOF Plus paperADTOF ISMIR 2021 paperPinned ADTOF Plus technical guideOriginal ADTOF sourceISMIR 2021 paper materialsSignals 2023 paperADTOF-PyTorch READMEADTOF (original)

ADTOF Plus

ADTOF Plus source (pinned)ADTOF PyTorch source (pinned)ADTOF Plus paper ADTOF ISMIR 2021 paper Pinned ADTOF Plus technical guide

Official ADTOF Plus input_is_mix control. Turn it on for a complete mix so MDX23C separation runs; turn it off only when the upload is already an isolated drum track.
Optional official ADTOF Plus default_threshold. Leave blank to keep the upstream -Infinity default; ADTOF PyTorch Direct does not use it.
After inference, Beat This final0 writes a valid fixed tempo or beat-by-beat tempo map into the MIDI while preserving every event's position in seconds. If valid beat evidence cannot be produced, the task fails instead of delivering MIDI with the model's default tempo.

ADTOF Direct · PyTorch

ADTOF PyTorch source (pinned)Original ADTOF source ISMIR 2021 paper materials Signals 2023 paper

Five separate official ADTOF thresholds for kick, snare, hi-hat, tom and cymbal; no comma-separated text entry.
Advanced control from the official ADTOF CLI. The upstream default is 100 FPS; changing it alters time resolution and may affect quality.
After inference, Beat This final0 writes a valid fixed tempo or beat-by-beat tempo map into the MIDI while preserving every event's position in seconds. If valid beat evidence cannot be produced, the task fails instead of delivering MIDI with the model's default tempo.

Paper and benchmark notes

Enhanced Automatic Drum Transcription via Drum Stem Source Separation

arXiv - 2025

Paper Pinned implementation

ADTOF: A large dataset of non-synthetic music for automatic drum transcription

ISMIR - 2021

Paper Paper materials

High-Quality and Reproducible Automatic Drum Transcription from Crowdsourced Data

Signals - 2023

Paper

Official / repositories

ADTOF Plus source (pinned)ADTOF PyTorch source (pinned)Original ADTOF source

Papers / technical notes

ADTOF Plus paper ADTOF ISMIR 2021 paper Pinned ADTOF Plus technical guide ISMIR 2021 paper materials Signals 2023 paper

Parameter guide

Official ADTOF Plus input_is_mix control. Turn it on for a complete mix so MDX23C separation runs; turn it off only when the upload is already an isolated drum track.
Optional official ADTOF Plus default_threshold. Leave blank to keep the upstream -Infinity default; ADTOF PyTorch Direct does not use it.
Five separate official ADTOF thresholds for kick, snare, hi-hat, tom and cymbal; no comma-separated text entry.
Advanced control from the official ADTOF CLI. The upstream default is 100 FPS; changing it alters time resolution and may affect quality.
After inference, Beat This final0 writes a valid fixed tempo or beat-by-beat tempo map into the MIDI while preserving every event's position in seconds. If valid beat evidence cannot be produced, the task fails instead of delivering MIDI with the model's default tempo.

Workflow

1Choose ADTOF Plus for a full song or Direct for an isolated drum track.
2Download drum MIDI and review kick, snare, hi-hat and other kit parts in a drum editor.

Before you submit

Full mix runs the ADTOF Plus separation workflow; Drums only bypasses separation and expects an isolated drum stem.
Advanced control from the official ADTOF CLI. The upstream default is 100 FPS; changing it alters time resolution and may affect quality.

Related use-case guides

Working on a similar task? These guides list the matching tools, model boundaries, sources, and FAQs for each scenario.

Audio to MIDI · online tool and model guide

Quick feedback

Include the tool name, steps and the problem in your feedback. Screenshots can help us locate it.