Computational auditory scene analysis

By Guy J. Brown, Martin P. Cooke. 1994

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

1. Deliang Wang
2. Dong Yu
3. Yoshua Bengio
4. Jinyu Li
5. Li Deng
6. Shinji Watanabe
7. Zhuo Chen
8. Kyunghyun Cho

Citations in the corpus

• Monaural speech segregation based on pitch tracking and amplitude modulation (2004, methodology): “Research in ASA has inspired considerable work to build computational auditory scene analysis (CASA) systems for sound segregation [5], [8], [12], [29], [33], [34].”
• Learning Spectral Clustering, With Application To Speech Separation (2006, background): “…and image segmentation, databases of hand-labelled segmented images are now available (Martin et al., 2001), while for the blind separation of speech signals via partitioning of the time-frequency plane (Brown and Cooke, 1994), training examples can be created by mixing previously captured signals.”
• Spectral Clustering for Speech Separation (2009, background): “…and image segmentation, databases of hand-labeled segmented images are now available (Martin et al., 2001), while for the blind separation of speech signals via partitioning of the time-frequency plane (Brown and Cooke, 1994), training examples can be created by mixing previously captured signals.”
• A real-time music-scene-description system: predominant-F0 estimation for detecting melody and bass lines in real-world audio signals (2004, background): “All rights reserved. doi:10.1016/j.specom.2004.07.001 Rosenthal and Okuno, 1998) is sound source segregation: the extraction of the audio signal corresponding to each auditory stream in a sound mixture.”
• SpeakerBeam: Speaker Aware Neural Network for Target Speaker Extraction in Speech Mixtures (2019, methodology): “This problem has been studied in the past using methods, such as Computational auditory scene analysis [3], [4], Non-negative matrix factorization [5], [6]”
• A Tandem Algorithm for Pitch Estimation and Voiced Speech Segregation (2010, methodology): “Voiced speech is characterized by periodicity (or har-monicity), which has been used as a primary cue in many CASA systems for segregating voiced speech (e.g., [8] and [16] ).”
• BodyScope: a wearable acoustic sensor for activity recognition (2012, background): “Most related to activity recognition is computational auditory scene recognition (CASR) [26], which differs from computational auditory scene analysis (CASA) [6].”
• Multiple fundamental frequency estimation based on harmonicity and spectral smoothness (2003, methodology): “More recent transcription systems have recruited psychoa-coustically motivated analysis principles, used sophisticated processing architectures, and extended the application area to computational auditory scene analysis in general [12].”
• Single-channel signal separation using time-domain basis functions (2003, background): “It is very difficult to compare a separation method with other CASA techniques, because their approaches are so different in many ways that an optimal tuning of their parameters would be beyond the scope of this letter.”
• Learning Spectral Clustering (2003, unknown): “features that are available in these domains [6, 7].”
• A Supervised Learning Approach to Monaural Segregation of Reverberant Speech (2009, background): “In segmentation, the input is decomposed into T-F segments, each of which is a contiguous region and deemed to mainly originate from a single source; in grouping, those segments that likely come from the same source are grouped into a stream by using pitch and other grouping cues.”
• A multi-pitch tracking algorithm for noisy speech (2002, methodology): “For example, if the background contains harmonic structures such as background music or voiced speech, a multipitch tracker would be required for providing meaningful pitch tracks.”
• Cracking the cocktail party problem by multi-beam deep attractor network (2017, background): “The cocktail party problem has been one of the most difficult challenges in audio signal processing for more than 60 years [1, 2, 3].”
• Robust automatic speech recognition with missing and unreliable acoustic data (2001, methodology): “There are alternative solutions to problem (i) based, for instance, on Auditory Scene Analysis (Cooke 1993, Brown and Cooke 1994, Ellis 1996).”
• Continuous latent variable models for dimensionality reduction and sequential data reconstruction (2001, methodology): “Perhaps perceptual grouping based on Gestalt principles, as used in computational auditory scene analysis (Brown and Cooke, 1994; Cooke and Ellis, 2000), could be helpful here.”
• Soft Mask Methods for Single-Channel Speaker Separation (2007, background): “Brown and Cooke [7] derive spectrographic masks through a detailed model that utilizes various cues such as periodicities in the signal, onsets and offsets, temporal continuity, and a variety of other physiologically motivated factors.”