# ByteCover: Cover Song Identification via Multi-Loss Training

arXiv: https://arxiv.org/abs/2010.14022 · HTML: https://mlsanity.com/paper/2010.14022/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 27 Oct 2020.

## Built on by (most cited first, 9 of 9)

- [CoverHunter: Cover Song Identification with Refined Attention and Alignments](https://mlsanity.com/paper/2306.09025/) (2023)
- [Semi-supervised music emotion recognition using noisy student training and harmonic pitch class profiles](https://mlsanity.com/paper/2112.00702/) (2021)
- [Bob's Confetti: Phonetic Memorization Attacks in Music and Video Generation](https://mlsanity.com/paper/2507.17937/) (2025)
- [Refining music sample identification with a self-supervised graph neural network](https://mlsanity.com/paper/2506.14684/) (2025)
- [MAIA: An Inpainting-Based Approach for Music Adversarial Attacks](https://mlsanity.com/paper/2509.04980/) (2025)
- [Innovations in Cover Song Detection: A Lyrics-Based Approach](https://mlsanity.com/paper/2406.04384/) (2024)
- [A Semi-Supervised Deep Learning Approach to Dataset Collection for Query-By-Humming Task](https://mlsanity.com/paper/2312.01092/) (2023)
- [Scalable Music Cover Retrieval Using Lyrics-Aligned Audio Embeddings](https://mlsanity.com/paper/2601.11262/) (2026)
- [Quality Audio Prototyping: a prototype system for unified sound retrieval and procedural generation](https://mlsanity.com/paper/2606.00629/) (2026)

## Abstract (arXiv)

We present in this paper ByteCover, which is a new feature learning method for cover song identification (CSI). ByteCover is built based on the classical ResNet model, and two major improvements are designed to further enhance the capability of the model for CSI. In the first improvement, we introduce the integration of instance normalization (IN) and batch normalization (BN) to build IBN blocks, which are major components of our ResNet-IBN model. With the help of the IBN blocks, our CSI model can learn features that are invariant to the changes of musical attributes such as key, tempo, timbre and genre, while preserving the version information. In the second improvement, we employ the BNNeck method to allow a multi-loss training and encourage our method to jointly optimize a classification loss and a triplet loss, and by this means, the inter-class discrimination and intra-class compactness of cover songs, can be ensured at the same time. A set of experiments demonstrated the effectiveness and efficiency of ByteCover on multiple datasets, and in the Da-TACOS dataset, ByteCover outperformed the best competitive system by 20.9%.