# Feature Learning for Chord Recognition: The Deep Chroma Extractor

arXiv: https://arxiv.org/abs/1612.05065 · HTML: https://mlsanity.com/paper/1612.05065/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 15 Dec 2016.

## Built on by (most cited first, 8 of 8)

- [Conditional Vector Graphics Generation for Music Cover Images](https://mlsanity.com/paper/2205.07301/) (2022)
- [Audio feature ranking for sound-based COVID-19 patient detection](https://mlsanity.com/paper/2104.07128/) (2021)
- [ChordFormer: A Conformer-Based Architecture for Large-Vocabulary Audio Chord Recognition](https://mlsanity.com/paper/2502.11840/) (2025)
- [Training chord recognition models on artificially generated audio](https://mlsanity.com/paper/2508.05878/) (2025)
- [Joint Transcription of Acoustic Guitar Strumming Directions and Chords](https://mlsanity.com/paper/2508.07973/) (2025)
- [Transcribing Rhythmic Patterns of the Guitar Track in Polyphonic Music](https://mlsanity.com/paper/2510.05756/) (2025)
- [MF-GCN: A Multi-Frequency Graph Convolutional Network for Tri-Modal Depression Detection Using Eye-Tracking, Facial, and Acoustic Features](https://mlsanity.com/paper/2511.15675/) (2025)
- [Dissonance Spectrum explicitly models perceptual frequency interactions for better music understanding](https://mlsanity.com/paper/2608.25621/) (2026)

## Abstract (arXiv)

We explore frame-level audio feature learning for chord recognition using artificial neural networks. We present the argument that chroma vectors potentially hold enough information to model harmonic content of audio for chord recognition, but that standard chroma extractors compute too noisy features. This leads us to propose a learned chroma feature extractor based on artificial neural networks. It is trained to compute chroma features that encode harmonic information important for chord recognition, while being robust to irrelevant interferences. We achieve this by feeding the network an audio spectrum with context instead of a single frame as input. This way, the network can learn to selectively compensate noise and resolve harmonic ambiguities. We compare the resulting features to hand-crafted ones by using a simple linear frame-wise classifier for chord recognition on various data sets. The results show that the learned feature extractor produces superior chroma vectors for chord recognition.