Razvanix/music-transcription-model

Music Transcription Model

This model performs automatic music transcription, converting audio recordings to MIDI notes.

Model Description

• Developed by: Razvan Calauz
• Model type: Audio-to-MIDI transcription
• Language(s): N/A (Audio processing)
• License: MIT
• Framework: TensorFlow 2.15.0

Intended Use

This model is designed to transcribe musical audio recordings into MIDI format for educational and research purposes.

How to Use

import tensorflow as tf
from huggingface_hub import snapshot_download

# Download model
model_path = snapshot_download(repo_id="Razvanix/music-transcription-model")

# Load model
model = tf.saved_model.load(model_path)