# Audio to Multitrack MIDI (Draft)

Split a full mix into stems (drums/bass/other/vocals) and transcribe each to MIDI — a multi-track starting point for transcription

> Canonical page: https://test.elysiatools.com/en/tools/audio-to-multitrack-midi

- **Category:** Media

- **Keywords:** audio, midi, transcription, multitrack, demucs, basic-pitch, stems, music, sheet, 扒谱

## Overview

Testing only: this resource-intensive tool accepts audio clips up to 20 seconds for every stem selection. It turns a full mix into a multi-track MIDI draft using two neural networks: HT-Demucs FT (ONNX) separates the audio into drums, bass, other instruments, and vocals; then Spotify's Basic Pitch (TensorFlow.js) transcribes each stem into MIDI notes. The stems are assembled into a single .mid file with one track per instrument. This is a draft / starting point for human transcription, not a perfect transcription. Expect missing notes, timing drift, and instrument bleed. Import the result into a DAW and clean it up by hand. Runs entirely in Node (no Python, no GPU).

## Inputs

- **Audio File** (file): Upload an MP3, WAV, FLAC, M4A, OGG, or AAC file (testing only; max 20 seconds)
- **Stems to Transcribe** (select)
- **Onset Sensitivity** (number)
- **Frame Sensitivity** (number)
- **Min Note Length (frames)** (number)
- **Include Pitch Bend** (checkbox)

## When to use

- Create a starting MIDI transcription from a short full-mix recording.
- Extract a vocal melody or other melodic material as MIDI notes.
- Separate drums, bass, other instruments, and vocals into tracks for manual editing.

## How it works

- Upload one MP3, WAV, FLAC, M4A, OGG, or AAC audio file up to 20 seconds.
- Choose which stems to transcribe: all, vocals only, melody, or full minus drums.
- Adjust onset sensitivity, frame sensitivity, minimum note length, and pitch-bend inclusion if needed.
- The tool separates the audio, transcribes the selected stems, and returns one .mid file with a track for each instrument.

## Use cases

- Music producers can turn a short mix into a multitrack MIDI starting point for arrangement and editing.
- Songwriters can extract a vocal line or melody as MIDI for further composition work.
- Musicians transcribing recordings can use separated drum, bass, instrument, and vocal tracks to speed up manual notation.

## Frequently asked questions

### What audio formats can I upload?

You can upload MP3, WAV, FLAC, M4A, OGG, or AAC files.

### What is the maximum audio length?

Audio clips are limited to 20 seconds.

### Which stems can the tool transcribe?

You can select all stems, vocals only, melody, or full mix minus drums.

### Is the MIDI transcription exact?

No. It is a draft and may contain missing notes, timing drift, and instrument bleed.

### What should I do with the MIDI output?

Import the .mid file into a DAW and correct the notes and timing manually.

## Related tools

- [Audio Melody Contour Extractor](https://test.elysiatools.com/en/tools/audio-melody-contour-extractor): Extract a dominant melody path from audio and export MIDI, note events, a pitch contour, SVG, and JSON in one ZIP.
- [Batch Audio Converter](https://test.elysiatools.com/en/tools/audio-batch-converter): Convert multiple audio files between different formats like MP3, WAV, FLAC, AAC, OGG, Opus with quality settings and batch processing
- [Audio Codec Swap](https://test.elysiatools.com/en/tools/audio-codec-swap): Change the codec of an audio file within a chosen output format
- [Audio Chord Progression Detector](https://test.elysiatools.com/en/tools/audio-chord-progression-detector): Estimate an audio file's chord sequence locally and download CSV, JSON, SVG, and method notes in one ZIP.
- [Image Particle Music Visualizer](https://test.elysiatools.com/en/tools/image-particle-music-visualizer): Turn a picture into an audio-reactive point cloud. Preview locally, then export a 30 fps MP4 with music, a PNG poster and settings JSON.
- [Audio to Text Transcriber (AI)](https://test.elysiatools.com/en/tools/audio-to-text-transcriber): Transcribe speech from audio (wav/mp3/m4a/flac/ogg/webm/aac) to text, SRT, VTT or JSON using the grok-stt AI model. Up to 10 minutes.
- [Audio Format Converter](https://test.elysiatools.com/en/tools/audio-converter): Convert audio files between different formats like MP3, WAV, FLAC, AAC, OGG with quality settings and volume adjustment
- [Hearing Safety Checker](https://test.elysiatools.com/en/tools/audio-hearing-safety-checker): Estimate your listening-exposure risk against NIOSH (85 dBA/8h, 3 dB exchange), OSHA (90 dBA/8h, 5 dB) and WHO-ITU H.870 weekly allowance — enter level, duration and sessions, or upload the audio you listen to for measured duration and loudness.

## Samples

- [Copyright-Free FLAC Audio Samples](https://test.elysiatools.com/en/samples/flac-samples): Lossless FLAC audio samples for testing and development, mirrored from MP3 set with nature sounds and meditation music
- [Copyright-Free MP3 Audio Samples](https://test.elysiatools.com/en/samples/mp3-samples): Collection of royalty-free audio samples for testing and development purposes including nature sounds, meditation music, and ambient audio
- [Copyright-Free WAV Audio Samples](https://test.elysiatools.com/en/samples/wav-samples): Uncompressed PCM WAV audio samples for testing and development, mirrored from MP3 set with nature sounds and meditation music
- [Copyright-Free M4A Audio Samples](https://test.elysiatools.com/en/samples/m4a-samples): Apple iTunes compatible M4A audio samples with AAC encoding, optimized for quality and file size balance

## Related content

- [Karaoke and Stem Separation Tools](https://test.elysiatools.com/en/hubs/karaoke-and-stem-separation-tools): Separate vocals and accompaniment, clean the resulting stems, set a controlled balance, and prepare a karaoke or remix-ready delivery.
