Acoustic classification of multiple simultaneous bird species: a multi-instance multi-label approach.

  title={Acoustic classification of multiple simultaneous bird species: a multi-instance multi-label approach.},
  author={Forrest Briggs and Balaji Lakshminarayanan and Lawrence Neal and Xiaoli Z. Fern and Raviv Raich and Sarah Hadley and Adam S. Hadley and Matthew G. Betts},
  journal={The Journal of the Acoustical Society of America},
  volume={131 6},
Although field-collected recordings typically contain multiple simultaneously vocalizing birds of different species, acoustic species classification in this setting has received little study so far. This work formulates the problem of classifying the set of species present in an audio recording using the multi-instance multi-label (MIML) framework for machine learning, and proposes a MIML bag generator for audio, i.e., an algorithm which transforms an input audio signal into a bag-of-instances… 

Figures and Tables from this paper

The 9th annual MLSP competition: New methods for acoustic classification of multiple simultaneous bird species in a noisy environment

It is an open problem for signal processing and machine learning to reliably identify bird sounds in real-world audio data collected in an acoustic monitoring scenario.

Acoustic Recognition of Multiple Bird Species Based on Penalized Maximum Likelihood

Time-frequency segmentation of the acoustic scene is obtained by employing a sinusoidal detection algorithm, which does not require any estimate of noise and is able to handle multiple simultaneous bird vocalizations.

Deep learning for detection of bird vocalisations

A deep autoencoder is presented that maps the audio spectrogram of bird vocalizations to its corresponding binary mask that encircles the spectral blobs of vocalizations while suppressing other audio sources.

Multi-Label Classifier Chains for Bird Sound

This work proposes to use an ensemble of classifier chains combined with a histogram-of-segments representation for multi-label classification of birdsong, and shows that the proposed method usually outperforms binary relevance and is better in some cases and worse in others compared to the MIML algorithms.

Deep Multiple Instance Learning Ensemble for the Acoustic Detection of Tropical Birds

This work proposes an ensemble of deep neural networks based on the MIL framework to predict the presence or absence of tropical birds in one-minute recordings and compares the performance of the individual networks using log mel-scaled spectrogram and mel-frequency cepstral coefficients.

Multi-Label Classification of Frog Species via Deep Learning

The results show that the proposed features extracted via pre-trained networks can achieve better classification performance when compared to hand-crafted features for classifying multiple simultaneously vocalizing species.

Multiple Instance Learning-Based Birdsong Classification Using Unsupervised Recording Segmentation

The proposed unsupervised segmentation method has a practical advantage over the compared supervised method, which requires the training from manually segmented spectrograms and is shown that the proposed approach is feasible to design an automatic recognition system of recordings which only requires, as training information, labeled examples of audio recordings.

Multi-Instance Multi-Label Class Discovery: A Computational Approach for Assessing Bird Biodiversity

This work employs a multi-instance multi-label representation to address multiple simultaneously vocalizing birds with sounds that overlap in time, and proposes new algorithms for species/class discovery using this representation.

Time-frequency segmentation of bird song in noisy acoustic environments

A supervised time-frequency audio segmentation method using a Random Forest classifier, to extract syllables of bird call from a noisy signal, outperforming energy thresholding.

Automated identification of field-recorded songs of four British grasshoppers using bioacoustic signal recognition

The system described here achieves automated discrimination between different species by utilizing a novel time domain signal coding technique and an artificial neural network and shows that it is capable of not only correctly recognizing the target species under a range of acoustic conditions but also of recognizing other sounds such as birds and man-made sounds.

Automated recognition of bird song elements from continuous recordings using dynamic time warping and hidden Markov models: a comparative study.

The performance of two techniques is compared for automated recognition of bird song units from continuous recordings, and one gives excellent to satisfactory performance and the other requires careful selection of templates.

Semi-automatic classification of bird vocalizations using spectral peak tracks.

An experiment using computer software to perform peak tracking of spectral analysis data demonstrates the usefulness of the sum-of-sinusoids model for rapid automatic recognition of isolated bird syllables.

Automatic Detection and Recognition of Tonal Bird Sounds in Noisy Environments

The proposed detection method shows high detection accuracy of bird tonal components and significant recognition accuracy improvements over the Mel-frequency cepstral coefficients, in both standard and noise-compensated models, and strong robustness to mismatch between the training and testing conditions.

Parametric Representations of Bird Sounds for Automatic Species Recognition

Three different parametric representations based on sinusoidal modeling,Mel-cepstrum parameters, and a vector of various descriptive features are tested because such models are popular in audio classification applications, and bird song is almost like music.

Birdsong recognition using backpropagation and multivariate statistics

An investigation has been made of bird species recognition using recordings of birdsong, with discernible recording noise such as tape hiss and, in some cases, other competing songs in the background.

Bird classification algorithms: theory and experimental results

  • C. KwanG. Mei K. Ho
  • Computer Science
    2004 IEEE International Conference on Acoustics, Speech, and Signal Processing
  • 2004
This paper summarizes the development of a high performance bird classification system using a hidden Markov model (HMM) and Gaussian mixture model (GMM).

Automatic Recognition of Bird Songs Using Cepstral Coefficients

A method to automatically identify birds from the sounds they generate by segmenting vocalization features and exploiting Linear discriminant analysis (LDA) to increase the classification accuracy at a lower dimensional feature vector space.