The bag-of-frames approach: a not so sufficient model for urban soundscapes

The "bag-of-frames" (BOF) approach, which encodes audio signals as the long-term statistical distribution of short-term spectral features, is commonly regarded as an effective and sufficient way to represent environmental sound recordings (soundscapes). The present paper describes a conceptual replication of a use of the BOF approach in a seminal article using several other soundscape datasets, with results strongly questioning the adequacy of the BOF approach for the task. As demonstrated in this paper, the good accuracy originally reported with BOF likely resulted from a particularly permissive dataset with low within-class variability. Soundscape modeling, therefore, may not be the closed case it was once thought to be.

• Publication year

  2014

• Venue

  Journal of the Acoustical Society of America

• Publication date

  2014-12-11

• Fields of study

  Physics, Computer Science, Mathematics, Engineering, Environmental Science, Medicine

• Identifiers
  DOI 10.1121/1.4935350arXiv 1412.4052PMID 26627819
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar, PubMed

• No claims are published for this paper.

• No concepts are published for this paper.

• Acoustic Scene Classification

  2015cited by this paper
• A Systematic Evaluation of the Bag-of-Frames Representation for Music Information Retrieval

  2014cited by this paper
• Sensing Urban Soundscapes

  2014cited by this paper
• Acoustic Scene Classification: Classifying environments from the sounds they produce

  2014influential reference
• To J.S.

  2014cited by this paper
• Capturing the workflows of music information retrieval for repeatability and reuse

  2013cited by this paper
• Detection and classification of acoustic scenes and events: An IEEE AASP challenge

  2013influential reference
• A database and challenge for acoustic scene classification and event detection

  2013influential reference
• Summary statistics in auditory perception

  2013cited by this paper
• An ear for statistics

  2013cited by this paper
• An analysis of the GTZAN music genre dataset

  2012cited by this paper
• Classification Accuracy Is Not Enough

  2012cited by this paper
• Informational factors in identifying environmental sounds in natural auditory scenes.

  2009cited by this paper
• Categorization of environmental sounds

  2009cited by this paper
• Judging the similarity of soundscapes does not require categorization: evidence from spliced stimuli.

  2009influential reference
• Perceptual study of soundscapes in train stations

  2008influential reference
• Categorization of environmental sounds.

  2007cited by this paper
• A Closer Look on Artist Filters for Musical Genre Classification

  2007cited by this paper
• The bag-of-frames approach to audio pattern recognition: a sufficient model for urban soundscapes but not for polyphonic music.

  2007influential reference
• Ecological validity of soundscape reproduction

  2004cited by this paper
• Effects of Context on the Identification of Everyday Sounds

  1991cited by this paper

• Polyphonic training set synthesis improves self-supervised urban sound classification.

  2021cites this paper
• Spectral images based environmental sound classification using CNN with meaningful data augmentation

  2021cites this paper
• SURROUNDING AUDIO SCENE AND EVENT IDENTIFICATION

  2021cites this paper
• Two-level fusion-based acoustic scene classification

  2020cites this paper
• A deep learning approach to sound classification for film audio post-production

  2020cites this paper
• A Low-Cost Sound Event Detection and Identification System for Urban Environments

  2020cites this paper
• Analysis and classification of acoustic scenes with wavelet transform-based mel-scaled features

  2020cites this paper
• Generative Model Driven Representation Learning in a Hybrid Framework for Environmental Audio Scene and Sound Event Recognition

  2020cites this paper
• Comparing Meta-Classifiers for Automatic Music Genre Classification

  2019cites this paper
• Texture Features for High-level Classification of Acoustic Scenes

  2019cites this paper
• Relevance-based quantization of scattering features for unsupervised mining of environmental audio

  2018influential citation
• Approaches to Complex Sound Scene Analysis

  2018cites this paper
• Classification of audio scenes with novel features in a fused system framework

  2018cites this paper
• SubSpectralNet – Using Sub-spectrogram Based Convolutional Neural Networks for Acoustic Scene Classification

  2018cites this paper
• Wavelet Transform Based Mel-scaled Features for Acoustic Scene Classification

  2018cites this paper
• Urban sound event classification based on local and global features aggregation

  2017cites this paper
• Environmental sound recognition using short-time feature aggregation

  2017cites this paper
• EigenScape : A Database of Spatial Acoustic Scene Recordings

  2017cites this paper
• ACOUSTIC SCENE CLASSIFICATION USING SPATIAL FEATURES

  2017cites this paper
• Automatic Detection of Screams and Shouts in the Metro

  2017cites this paper
• Soundscape Audio Signal Classification and Segmentation Using Listeners Perception of Background and Foreground Sound

  2016cites this paper
• On urban soundscape mapping : A computer can predict the outcome of soundscape assessments

  2016cites this paper
• Acoustic Scene Classification: An evaluation of an extremely compact feature representation

  2016cites this paper
• One hundred ways to process time, frequency, rate and scale in the central auditory system: a pattern-recognition meta-analysis

  2015cites this paper
• BF-Classifier: Background/Foreground Classification and Segmentation of Soundscape Recordings

  2015cites this paper
• Acoustic Scene Classification based on Sound Textures and Events

  2015cites this paper