piano-transcription-inference 0.0.3

Piano transcription inference toolbox

Piano transcription inference

This toolbox is a piano transcription inference package that can be easily installed. Users can transcribe their favorite piano recordings to MIDI files after installation. To see how the piano transcription system is trained, please visit: https://github.com/bytedance/piano_transcription.

Demos

Here is a demo of our piano transcription system: https://www.youtube.com/watch?v=5U-WL0QvKCg

Installation

The piano transcription system is developed with Python 3.7 and PyTorch 1.4.0 (Should work with other versions, but not fully tested). Install PyTorch (>=1.4) following https://pytorch.org/. Users should have ffmpeg installed to transcribe mp3 files.

pip install piano_transcription_inference

Then, installation finished!

Or users could also install from source code:

$ python3 setup.py install

Usage

python3 example.py --audio_path='resources/cut_liszt.mp3' --output_midi_path='cut_liszt.mid' --cuda

For example:

import librosa
from piano_transcription_inference import PianoTranscription, sample_rate

# Load audio
(audio, _) = librosa.core.load('resources/cut_liszt.mp3', sr=sample_rate, mono=True)

# Transcriptor
transcriptor = PianoTranscription(device='cuda')

# Transcribe and write out to MIDI file
transcribed_dict = transcriptor.transcribe(audio, 'cut_liszt.mid')

Visualization of piano transcription

Demo. Lang Lang: Franz Liszt - Love Dream (Liebestraum) [audio] [transcribed_midi]

FAQs

If users met the problem of command line shows "Killed". This can be caused by users do not have sufficient memory.

Applications

We have built a large-scale classical piano MIDI dataset https://github.com/bytedance/GiantMIDI-Piano using our piano transcription system.

Cite

[1] High-resolution Piano Transcription with Pedals by Regressing Onsets and Offsets Times, [To appear], 2020