# The impact of Audio input representations on neural network based music transcription

arXiv: https://arxiv.org/abs/2001.09989 · HTML: https://mlsanity.com/paper/2001.09989/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 25 Jan 2020.

## Built on by (most cited first, 4 of 4)

- [Video2Music: Suitable Music Generation from Videos using an Affective Multimodal Transformer model](https://mlsanity.com/paper/2311.00968/) (2023)
- [DiffRoll: Diffusion-based Generative Music Transcription with Unsupervised Pretraining Capability](https://mlsanity.com/paper/2210.05148/) (2022)
- [Transfer of knowledge among instruments in automatic music transcription](https://mlsanity.com/paper/2305.00426/) (2023)
- [Clean2FX: Label-conditioned modeling for clean-to-effect guitar audio transformations](https://mlsanity.com/paper/2607.08863/) (2026)

## Abstract (arXiv)

This paper thoroughly analyses the effect of different input representations on polyphonic multi-instrument music transcription. We use our own GPU based spectrogram extraction tool, nnAudio, to investigate the influence of using a linear-frequency spectrogram, log-frequency spectrogram, Mel spectrogram, and constant-Q transform (CQT). Our results show that a 8.33% increase in transcription accuracy and a 9.39% reduction in error can be obtained by choosing the appropriate input representation (log-frequency spectrogram with STFT window length 4,096 and 2,048 frequency bins in the spectrogram) without changing the neural network design (single layer fully connected). Our experiments also show that Mel spectrogram is a compact representation for which we can reduce the number of frequency bins to only 512 while still keeping a relatively high music transcription accuracy.