EfficientLEAF: A Faster LEarnable Audio Frontend of Questionable Use

07/12/2022
by   Jan Schlüter, et al.
0

In audio classification, differentiable auditory filterbanks with few parameters cover the middle ground between hard-coded spectrograms and raw audio. LEAF (arXiv:2101.08596), a Gabor-based filterbank combined with Per-Channel Energy Normalization (PCEN), has shown promising results, but is computationally expensive. With inhomogeneous convolution kernel sizes and strides, and by replacing PCEN with better parallelizable operations, we can reach similar results more efficiently. In experiments on six audio classification tasks, our frontend matches the accuracy of LEAF at 3 cost, but both fail to consistently outperform a fixed mel filterbank. The quest for learnable audio frontends is not solved.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
01/21/2021

LEAF: A Learnable Frontend for Audio Classification

Mel-filterbanks are fixed, engineered audio features which emulate human...
research
04/30/2023

Transformer-based Sequence Labeling for Audio Classification based on MFCCs

Audio classification is vital in areas such as speech and music recognit...
research
10/03/2022

Simple Pooling Front-ends For Efficient Audio Classification

Recently, there has been increasing interest in building efficient audio...
research
02/13/2019

Improving performance and inference on audio classification tasks using capsule networks

Classification of audio samples is an important part of many auditory sy...
research
02/22/2020

Multi-Representation Knowledge Distillation For Audio Classification

As an important component of multimedia analysis tasks, audio classifica...
research
10/06/2021

An Investigation of the Effectiveness of Phase for Audio Classification

While log-amplitude mel-spectrogram has widely been used as the feature ...
research
08/05/2022

Deep Feature Learning for Medical Acoustics

The purpose of this paper is to compare different learnable frontends in...

Please sign up or login with your details

Forgot password? Click here to reset