Computational Bioacoustics

Computational Bioacoustics

by Todor Ganchev
Released June 2017
Publisher(s): De Gruyter
ISBN: 9781614519669

Read it now on the O’Reilly learning platform with a 10-day free trial.

O’Reilly members get unlimited access to books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.

Start your free trial

Book description

This book offers an overview of some recent advances in the Computational Bioacoustics methods and technology. In the focus of discussion is the pursuit of scalability, which would facilitate real-world applications of different scope and purpose, such as wildlife monitoring, biodiversity assessment, pest population control, and monitoring the spread of disease transmitting mosquitoes. The various tasks of Computational Bioacoustics are described and a wide range of audio parameterization and recognition tasks related to the automated recognition of species and sound events is discussed. Many of the Computational Bioacoustics methods were originally developed for the needs of speech, audio, or image processing, and afterwards were adapted to the requirements of automated acoustic recognition of species, or were elaborated further to address the challenges of real-world operation in 24/7 mode. The interested reader is encouraged to follow the numerous references and links to web resources for further information and insights. This book is addressed to Software Engineers, IT experts, Computer Science researchers, Bioacousticians, and other practitioners concerned with the creation of new tools and services, aimed at enhancing the technological support to Computational Bioacoustics applications.

STTM, Speech Technology and Text Mining in Medicine and Health Care

This series demonstrates how the latest advances in speech technology and text mining positively affect patient healthcare and, in a much broader sense, public health at large. New developments in text mining methods have allowed health care providers to monitor a large population of patients at any time and from any location. Employing advanced summarization techniques, patient data can be readily extracted from extensive clinical documents in electronic health records and immediately made available to the physician. These same summarization techniques can also aid the healthcare provider in extracting from the large corpora of medical literature the relevant information for treating the patient. The series topics include the design and acceptance of speech-enabled robots that assist in the operating room, studies of signal processing and acoustic modeling for speech and communication disorders, advanced statistical speech enhancement methods for creating synthetic voice, and technologies for addressing speech and language impairments. Titles in the Series consist of both authored books and edited contributions. All authored books and contributed works are peer-reviewed. The Series is for speech scientists and speech engineers, machine learning experts, biomedical engineers, medical speech pathologists, linguists, and healthcare professionals

Table of contents

  1. Cover
  2. Title Page
  3. Copyright
  4. Abstract
  5. Preface
  6. Contents
  7. Acronyms
  8. 1 Introduction
    1. 1.1 Why should we care about biodiversity?
    2. 1.2 Role of computational bioacoustics
    3. 1.3 The benefits of computational bioacoustics
    4. 1.4 Book organization
    5. 1.5 How to use this book
    6. 1.6 The home page for this book
  9. 2 Why computational bioacoustics?
    1. Introduction
    2. 2.1 Ancient roots
    3. 2.2 Contemporary bioacoustics
    4. 2.3 Computational bioacoustics
    5. 2.4 Relations between bioacoustics and computational bioacoustics
    6. 2.5 Research topics and application areas
  10. 3 Definition of tasks
    1. Introduction
    2. 3.1 One-species detection
    3. 3.2 Species identification
    4. 3.3 Multi-label species identification
    5. 3.4 One-category recognition
    6. 3.5 One-species recognition
    7. 3.6 Multispecies diarization
    8. 3.7 Localization and tracking of individuals
    9. 3.8 Sound event-type recognition
    10. 3.9 Abundance assessment
    11. 3.10 Clustering of sound events
    12. 3.11 Biodiversity assessment indices
    13. 3.12 Practical difficulties related to real-world operation
  11. 4 Acoustic libraries
    1. Introduction
    2. 4.1 The training library
      1. 4.1.1 TL organization
      2. 4.1.2 Creation of training libraries
      3. 4.1.3 Manual creation of TLs
      4. 4.1.4 TLs based on clean recordings
      5. 4.1.5 Challenges related to TLs’ creation
      6. 4.1.6 Computational bioacoustics in support of TL creation
      7. 4.1.7 What if large TLs are not feasible?
    3. 4.2 The acoustic background library
      1. 4.2.1 What if large BL is not available?
      2. 4.2.2 Computational bioacoustics in support of BL creation
    4. 4.3 The validation library
      1. 4.3.1 Computational bioacoustics in support of VL creation
    5. 4.4 The evaluation library
      1. 4.4.1 Computational bioacoustics in support of EL creation
    6. Concluding remarks
  12. 5 One-dimensional audio parameterization
    1. Introduction
    2. 5.1 Audio pre-processing
      1. 5.1.1 Variable-length segmentation
      2. 5.1.2 Uniform-length segmentation
    3. 5.2 Audio parameterization
    4. 5.3 Post-processing of audio features
      1. 5.3.1 Statistical standardization
      2. 5.3.2 Temporal derivatives
      3. 5.3.3 Shifted delta coefficients
    5. Concluding remarks
  13. 6 Two-dimensional audio parameterization
    1. Introduction
    2. 6.1 The audio spectrogram
    3. 6.2 The thresholded spectrogram
    4. 6.3 Morphological filtering of the thresholded spectrogram
    5. 6.4 MFCC estimation with robust frames selection
    6. 6.5 Points of interest-based features
    7. 6.6 Bag-of-instances audio descriptors
    8. Concluding remarks
  14. 7 Audio recognizers
    1. Introduction
    2. 7.1 Overview of classification methods
    3. 7.2 One-species detectors
    4. 7.3 Single-label classification with a single model per category
    5. 7.4 Single-label classification with multi-classifier schemes
    6. 7.5 MIML classification schemes
    7. Concluding remarks
  15. 8 Application examples
    1. Introduction
    2. 8.1 The ARBIMON project
      1. 8.1.1 The ARBIMON Acoustics project
      2. 8.1.2 The ARBIMON II project
    3. 8.2 The AmiBio project
    4. 8.3 The Pantanal Biomonitoring Project
    5. 8.4 The SABIOD project
    6. 8.5 The ENTOMATIC project
    7. 8.6 The REMOSIS project
    8. Concluding remarks
  16. 9 Useful resources
    1. Introduction
    2. 9.1 Online audio data sets and resources
      1. 9.1.1 The xeno-canto repository
      2. 9.1.2 The Borror Laboratory of Bioacoustics Sound Archive
      3. 9.1.3 The Macaulay Library
      4. 9.1.4 The British Library Sound Archive
      5. 9.1.5 Sound Library of the Wildlife Sound Recording Society
      6. 9.1.6 The BioAcoustica repository
      7. 9.1.7 The Animal Sound Archive at the Museum für Naturkunde in Berlin
      8. 9.1.8 Australian National Wildlife Collection Sound Archive
      9. 9.1.9 The Western Soundscape Archive
      10. 9.1.10 The DORSA Archive
      11. 9.1.11 The Nature Sounds Society
      12. 9.1.12 The Acoustics Ecology Institute
      13. 9.1.13 Sound resources not listed here
    3. 9.2 Software tools and services
    4. 9.3 Online information
    5. 9.4 Scientific forums
    6. 9.5 Technology evaluation campaigns
    7. Concluding remarks
  17. Epilogue
  18. References
  19. Index

Product information

  • Title: Computational Bioacoustics
  • Author(s): Todor Ganchev
  • Release date: June 2017
  • Publisher(s): De Gruyter
  • ISBN: 9781614519669