# CGCNN: Complex Gabor Convolutional Neural Network on raw speech

arXiv: https://arxiv.org/abs/2002.04569 · HTML: https://mlsanity.com/paper/2002.04569/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 11 Feb 2020.

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

- [LEAF: A Learnable Frontend for Audio Classification](https://mlsanity.com/paper/2101.08596/) (2021)
- [A multi-level interpretable sleep stage scoring system by infusing experts' knowledge into a deep network architecture](https://mlsanity.com/paper/2207.04585/) (2022)
- [TACNET: Temporal Audio Source Counting Network](https://mlsanity.com/paper/2311.02369/) (2023)
- [Learning neural audio features without supervision](https://mlsanity.com/paper/2203.15519/) (2022)
- [Time-Frequency Localization Using Deep Convolutional Maxout Neural Network in Persian Speech Recognition](https://mlsanity.com/paper/2108.03818/) (2021)
- [Complementing Handcrafted Features with Raw Waveform Using a Light-weight Auxiliary Model](https://mlsanity.com/paper/2109.02773/) (2021)
- [Phonetic Error Analysis of Raw Waveform Acoustic Models with Parametric and Non-Parametric CNNs](https://mlsanity.com/paper/2406.00898/) (2024)
- [Raw Audio Classification with Cosine Convolutional Neural Network (CosCovNN)](https://mlsanity.com/paper/2412.00312/) (2024)
- [Phonetic Error Analysis of Raw Waveform Acoustic Models](https://mlsanity.com/paper/2606.07030/) (2026)

## Abstract (arXiv)

Convolutional Neural Networks (CNN) have been used in Automatic Speech Recognition (ASR) to learn representations directly from the raw signal instead of hand-crafted acoustic features, providing a richer and lossless input signal. Recent researches propose to inject prior acoustic knowledge to the first convolutional layer by integrating the shape of the impulse responses in order to increase both the interpretability of the learnt acoustic model, and its performances. We propose to combine the complex Gabor filter with complex-valued deep neural networks to replace usual CNN weights kernels, to fully take advantage of its optimal time-frequency resolution and of the complex domain. The conducted experiments on the TIMIT phoneme recognition task shows that the proposed approach reaches top-of-the-line performances while remaining interpretable.