paper-with-me

Papers

EfficientLEAF: A Faster LEarnable Audio Frontend of Questionable Use

2022-07-12 · Jan Schlüter, Gerald Gutenbrunner

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% of the cost, but both fail to consistently outperform a fixed mel filterbank. The quest for learnable audio frontends is not solved.

📄 PDF Abstract BibTeX arXiv:2207.05508

Code (1)

cpjku/efficientleaf 공식 구현 pytorch

Tasks

Audio ClassificationClassificationInstrument RecognitionPitch ClassificationSpoken language identification

Methods 이 논문이 사용한 방법론

Dropout Dropout is a regularization technique for neural networks that drops a unit (along with connections) at training time with a specified probability $p$ (a common value is…
Average Pooling 설명 없음
EfficientNet 설명 없음
Convolution A convolution is a type of matrix operation, consisting of a kernel, a small matrix of weights, that slides over input data performing element-wise multiplication with the…

Similar Papers 제목 키워드 기반

A Universal Learnable Audio Frontend

2021-01-01 · ICLR 2021 1 · Neil Zeghidour, Olivier Teboul, Félix de Chaumont Quitry, Marco Tagliasacchi

Mel-filterbanks are fixed, engineered audio features which emulate human perception and have lived through the history of audio understanding up to today. However, their undeniable qualities are counterbalanced by the fu…

Audio Classification

LEAF: A Learnable Frontend for Audio Classification

2021-01-21 · Neil Zeghidour, Olivier Teboul, Félix de Chaumont Quitry, Marco Tagliasacchi

Mel-filterbanks are fixed, engineered audio features which emulate human perception and have been used through the history of audio understanding up to today. However, their undeniable qualities are counterbalanced by th…

Audio ClassificationClassificationGeneral Classification

Learnable Frontends that do not Learn: Quantifying Sensitivity to Filterbank Initialisation

2023-02-20 · Mark Anderson, Tomi Kinnunen, Naomi Harte

While much of modern speech and audio processing relies on deep neural networks trained using fixed audio representations, recent studies suggest great potential in acoustic frontends learnt jointly with a backend. In th…

Action DetectionActivity DetectionSensitivity

Deep Feature Learning for Medical Acoustics

2022-08-05 · Alessandro Maria Poirè, Federico Simonetta, Stavros Ntalampiras

The purpose of this paper is to compare different learnable frontends in medical acoustics tasks. A framework has been implemented to classify human respiratory sounds and heartbeats in two categories, i.e. healthy or af…

Audio Classification

Learning neural audio features without supervision

2022-03-29 · Sarthak Yadav, Neil Zeghidour

Deep audio classification, traditionally cast as training a deep neural network on top of mel-filterbanks in a supervised fashion, has recently benefited from two independent lines of work. The first one explores "learna…

Audio ClassificationSelf-Supervised Learning