Environmental sound processing and its applications

Miyazaki, Koichi; Toda, Tomoki; Hayashi, Tomoki; Takeda, Kazuya

doi:10.1002/tee.22868

Cited by 5 publications

(3 citation statements)

References 95 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Related Surveys and Differences. While several comprehensive surveys delve into the applications of deep learning for audio processing [11,28,53,54], including speech [25,26], music [55][56][57], and other categories [58,59], none concentrate on the advent and deployment of LLMs in this field. Numerous surveys exist that cover the vast landscape of LLMs, each focusing on specific aspects or applications.…”

Section: Speechmentioning

confidence: 99%

A survey on deep reinforcement learning for audio-based applications

Latif

Cuayáhuitl

Pervez³

et al. 2022

Artif Intell Rev

View full text Add to dashboard Cite

Deep reinforcement learning (DRL) is poised to revolutionise the field of artificial intelligence (AI) by endowing autonomous systems with high levels of understanding of the real world. Currently, deep learning (DL) is enabling DRL to effectively solve various intractable problems in various fields including computer vision, natural language processing, healthcare, robotics, to name a few. Most importantly, DRL algorithms are also being employed in audio signal processing to learn directly from speech, music and other sound signals in order to create audio-based autonomous systems that have many promising applications in the real world. In this article, we conduct a comprehensive survey on the progress of DRL in the audio domain by bringing together research studies across different but related areas in speech and music. We begin with an introduction to the general field of DL and reinforcement learning (RL), then progress to the main DRL methods and their applications in the audio domain. We conclude by presenting important challenges faced by audio-based DRL agents and by highlighting open areas for future research and investigation. The findings of this paper will guide researchers interested in DRL for the audio domain.

show abstract

Section: Speechmentioning

confidence: 99%

A survey on deep reinforcement learning for audio-based applications

Latif

Cuayáhuitl

Pervez³

et al. 2022

Artif Intell Rev

View full text Add to dashboard Cite

show abstract

“…(Park & Kaneko, 2000;Yamoaka et al, 2002; Metode akustik tomografi pesisir ini juga dapat digunakan untuk memantau kondisi fisik perairan (Park & Kaneko, 2000;Yamoaka et al, 2002;. Penelitian terkait kebisingan lingkungan merupakan tantangan yang unik yang menawarkan alternatif penyelesaian masalah terkait observasi kondisi lingkungan perairan dengan cara menganalisa propagasi gelombang dan pemrosesan sinyal (Miyazaki et al, 2019). Untuk ringkasnya, penelitian ini bertujuan untuk melakukan karakterisasi dari fenomena-fenomena kelautan melalui analisa data kebisingan lingkungan yang terekam sebagai sinyal ESN, melalui pemahaman fisika gelombang, fenomenafenomena ini dapat teridentifikasi kemunculanya dan berkaitan erat dengan perubahan lingkungan seperti hujan lokal, lalu lintas perairan maupun sinyal-sinyal bawah laut.…”

Section: Pendahuluanunclassified

Kajian Kebisingan Lingkungan pada Akustik Tomografi Pesisir (ATP) untuk Prediksi Arus Permukaan Lokal

Pianto¹,

Akbar²,

Dinanta³

et al. 2022

Jurnal Kelautan Nasional

View full text Add to dashboard Cite

Akustik Tomografi Pesisir (ATP) merupakan sebuah inovasi teknologi yang dapat digunakan untuk mendapatkan informasi tentang karakter fisik kelautan pada kurun waktu tertentu melalui media gelombang akustik. Pada sistem ATP, pembacaan transducer/sensor memiliki peran yang sangat penting, karena data dan informasi sangat bergantung pada kestabilan instrument oleh karenanya pemasangan alat menjadi sangat penting. disamping itu, langkah-langkah akusisi dan pengolahan data ATP wajib memiliki fundamental tentang fisika dan pemahaman tentang kelautan yang baik. Salah satu yang harus dipahami adalah karakter kebisingan lingkungan yang terekam sebagai Environmental Sound Noise (ESN). Data ini diperoleh dengan tujuan khusus untuk melihat karakter lingkungan dimana sensor ATP dipasang pada waktu rentang waktu tertentu, disamping itu data ini dapat digunakan juga sebagai instrument pemantauan perubahan kondisi lingkungan secara near-realtime. Data pada penelitian ini menggunakan data ATP [j1] yg di pasang disuar navigasi di wilayah perairan Senggigi, Lombok pada bulan Januari tahun 2021, dengan jumlah data 2724 data, yang terbagi kedalam empat (4) kuadran waktu, Analisa dengan menggunakan fourier transform (FFT) dan power spectral density (PSD) membuka informasi tentang kondisi arus permukaan. Berdasarkan hasil kajian dan analisis yang telah dilakukan didapat informasi pada saat kondisi hujan lebat sinyal ESN akan memiliki spike amplitudo dengan frekuensi yang yang rapat, kemudian sea-breeze setelah dilakukan signal processing berada pada nilai <12 Hz, surface wavepada 12-15 Hz dan underwater signal 15-32 Hz. Secara garis besar ATP dapat menjadi instrument untuk pengukuran kondisi perairan dan menjadi salah satu referensi untuk lalu lintas pelayaran.

show abstract

“…The groundwork for this advance was laid, at least in part, by developments in computing technologies and ML improvements in other domains such as computer vision and image processing. A few examples of acoustical ML problems include the recognition and interpretation of human speech (Graves et al, 2013;Hinton et al, 2012;Nassif et al, 2019), classification of animal vocalizations (Acevedo et al, 2009;Aide et al, 2013;Bermant et al, 2019;Hildebrand et al, 2019;Moln ar et al, 2008;Shiu et al, 2020;Vickers et al, 2021), classification of music genre (Costa et al, 2017;Fu et al, 2011;Tzanetakis and Cook, 2002), localization of sound sources (Adavanne et al, 2019;Ferguson et al, 2018;Niu et al, 2017a,b;Sun et al, 2018), inference from physical acoustic measurements (Abord an and Szab o, 2021; Bianco and Gerstoft, 2017;Li et al, 2021;Piccolo et al, 2019), and classification of acoustic scenes and events (Abeßer, 2020;Mesaros et al, 2018;Miyazaki et al, 2019).…”

Section: Introductionmentioning

confidence: 99%

Evaluating machine learning architectures for sound event detection for signals with variable signal-to-noise-ratios in the Beaufort Sea

Ibrahim,

Sagers,

Ballard

et al. 2023

The Journal of the Acoustical Society of America

View full text Add to dashboard Cite

This paper explores the challenging polyphonic sound event detection problem using machine learning architectures applied to data recorded in the Beaufort Sea during the Canada Basin Acoustic Propagation Experiment. Four candidate architectures were investigated and evaluated on nine classes of signals broadcast from moored sources that were recorded on a vertical line array of hydrophones over the course of the yearlong experiment. These signals represent a high degree of variability with respect to time-frequency characteristics, changes in signal-to-noise ratio (SNR) associated with varying signal levels as well as fluctuating ambient sound levels, and variable distributions, which resulted in class imbalances. Within this context, binary relevance, which decomposes the multi-label learning task into a number of independent binary learning tasks, was examined as an alternative to the conventional multi-label classification (MLC) approach. Binary relevance has several advantages, including flexible, lightweight model configurations that support faster model inference. In the experiments presented, binary relevance outperformed conventional MLC approach on classes with the most imbalance and lowest SNR. A deeper investigation of model performance as a function of SNR showed that binary relevance significantly improved recall within the low SNR range for all classes studied.

show abstract

Environmental sound processing and its applications

Cited by 5 publications

References 95 publications

A survey on deep reinforcement learning for audio-based applications

A survey on deep reinforcement learning for audio-based applications

Kajian Kebisingan Lingkungan pada Akustik Tomografi Pesisir (ATP) untuk Prediksi Arus Permukaan Lokal

Evaluating machine learning architectures for sound event detection for signals with variable signal-to-noise-ratios in the Beaufort Sea

Contact Info

Product

Resources

About