OMAR-RQ is trained with self-supervision via masked token classification methodologies using a large-scale dataset with over 330,000 hours of music audio. It offers powerful, multipurpose representations essential for advancing research in music information retrieval. The model achieves state-of-the-art performance among open self-supervised models across various tasks:
For development including pre-training your own models:
pip install -e .[train]
Inference
You can load an OMAR-RQ model by specifying its Hugging Face model ID:
python
1import torch
2from omar_rq import get_model
34# Embedding extraction example5x = torch.randn(1,16000*4).cpu()# Example: 4 seconds of mono audio at 16kHz67# Load a specific model, e.g., "mtg-upf/omar-rq-multifeature-25hz-fsq"8model_id ="mtg-upf/omar-rq-multifeature-25hz-fsq"9model = get_model(model_id=model_id, device="cpu")# Use "cuda" if a GPU is available1011# Extract embeddings from layer 612embeddings = model.extract_embeddings(x, layers=[6])1314# Use the `model.eps` field to compute timestamps for the extracted embeddings15timestamps = torch.arange(embeddings.shape[2])/ model.eps
1617print(f"Extracted embeddings shape: {embeddings.shape}")18print(f"First 5 timestamps: {timestamps[:5]}")
get_model reference:
Returns an OMAR-RQ Module from the provided model_id or config_file.
Args:
model_id (str): Hugging Face's Model ID or local path to the model
config_file (Path): Path to the model config of a trained model.
device (str): Device to use for the model. Defaults to "cpu".
quantization_targets (bool): If True, it will create the quantization
targets for SSL pre-training of the model. Defaults to False.
Output:
module: The model from the provided config file.
Module usage:
Args:
audio (torch.Tensor): 2D mono audio tensor (B, T'). Where B is
the batch size and T' is the number of samples.
layers (set): Set of layer indices to extract embeddings from.
By default, it extracts embeddings from the last layer (logits).
Output:
torch.Tensor: Extracted embeddings. The output tensor has shape
(L, B, T, C,) where L = len(layers), B is the batch size, T is
the number of output timestamps, and C = embedding dimension.
extract_embeddings reference:
Extract embeddings from an input audio batch.
Args:
audio (torch.Tensor): 2D mono audio tensor (B, T'). Where B is
the batch size and T' is the number of samples.
layers (set): Set of layer indices to extract embeddings from.
By default, it extracts embeddings from the last layer (logits).
Output:
torch.Tensor: Extracted embeddings. The output tensor has shape
(L, B, T, C,) where L = len(layers), B is the batch size, T is
the number of output timestamps, and C = embedding dimension.
Available Models
OMAR-RQ models are offered in different configurations, each with its own strengths and weaknesses. Models based on mel spectrogram (base and multicodebook) tend to perform better on semantic tasks such as auto-tagging, structure recognition, and difficulty estimation. On the other hand, multifeature-25hz-fsq offers the best performance in tonal and temporal tasks such as pitch and chord estimation, and beat tracking.
If you find this work useful, please cite the paper:
bibtex
1@article {alonso2025omarrq,
2 title={OMAR-RQ: Open Music Audio Representation Model Trained with Multi-Feature Masked Token Prediction},
3 author={Alonso-Jim\'enez, Pablo and Ramoneda, Pedro and Araz, R. Oguz and Poltronieri, Andrea and Bogdanov, Dmitry},
4 journal={arXiv preprint arXiv:2507.03482},
5 year={2025}
6}
