scholarly journals A simulated “cocktail party” with up to three sound sources

1996 ◽  
Vol 58 (7) ◽  
pp. 1026-1036 ◽  
Author(s):  
William A. Yost ◽  
Raymond H. Dye ◽  
Stanley Sheft
Keyword(s):  
Cocktail Party ◽  
Sound Sources

Divided auditory attention with up to three sound sources: A cocktail party

1994 ◽  
Vol 95 (5) ◽  
pp. 2916-2916 ◽  
Author(s):  
William A. Yost ◽  
Stanley Sheft ◽  
Raymond (Toby) Dye
Keyword(s):  
Auditory Attention ◽  
Cocktail Party ◽  
Sound Sources

Blind Source Separation Using a Neural Network

2008 ◽  
Author(s):  
Yukihito Niino ◽  
Toshihiko Shiraishi ◽  
Shin Morishita
Keyword(s):  
Neural Network ◽  
Blind Source Separation ◽  
Sound Source ◽  
Sound Pressure ◽  
Source Separation ◽  
Cocktail Party ◽  
Time Step ◽  
Sound Sources ◽  
The Neural Network ◽  
Cocktail Party Effect

Humans are able to well recognize mixtures of speech signals produced by two or more simultaneous speakers. This ability is known as cocktail party effect. To apply the cocktail party effect to engineering, we can construct novel systems of blind source separation such as current automatic speech recognition systems and active noise control systems under environment noises. A variety of methods have been developed to improve the performance of blind source separation in the presence of background noise or interfering speech. Considering blind source separation as the characteristics of human, artificial neural networks are suitable for it. In this paper, we proposed a method of blind source separation using a neural network. The present neural network can adaptively separate sound sources on training the internal parameters. The network was three-layered. Sound pressure was output from two sound sources and the mixed sound was measured with two microphones. The time history of microphone signals was input to the input layer of neural network. The two outputs of hidden layer were corresponding to the two sound pressure separated respectively. The two outputs of output layer were corresponding to the two microphone signals expected at next time step and compared with the actual microphone signals at next time step to train the neural network by a backpropagation method. In this procedure, the signal from each sound source was adaptively separated. There were two conditions of sound source, sinusoidal signals of 440 and 1000 Hz. In order to assess the performance of neural network numerically and experimentally, a basic independent component analysis (ICA) was conducted simultaneously. The results obtained are as follows. The performance of blind separation by the neural network was higher than the basic ICA. In addition, the neural network can successfully separate the sound source in spite of the position of sound sources.

Underwater Sound Filtering

2015 ◽  
Author(s):  
Ines Simic ◽  
Rutger van Aalst
Keyword(s):  
Blind Source Separation ◽  
Water Waves ◽  
Source Separation ◽  
The Other ◽  
Underwater Sound ◽  
Noisy Environment ◽  
Cocktail Party ◽  
Sound Sources ◽  
Open Sea ◽  
Filtering Problem

The cocktail party algorithm is one of the most widely used algorithms for source separation of sound. The algorithm aims to find an automated solution for a problem that everyone experiences regularly, namely how to make oneself heard in a noisy environment. The cocktail party algorithm picks up the sound from different microphones, and then applies smart filters once the system has determined which sounds originate from the same source. This problem also becomes topical when developing sensors based on passive sonar, for instance for autonomous aquatic drones who have to develop awareness of ships and other possible obstacles on a busy shipping lane. It is possible to deploy multiple hydrophones to localize sound sources under water, but the system will be hindered considerably by the sound that the drone itself makes, such as the sound produced by the propellers. This paper describes a possible solution to the underwater sound filtering problem, using Blind Source Separation. The problem regards splitting sound from a boat engine and the water waves to prove the possibility to extract one sound fragment from the other on the open sea. The illustrations shown further in the report are tests performed in MATLAB to prove the theory.

scholarly journals Schema learning for the cocktail party problem

2018 ◽  
Vol 115 (14) ◽  
pp. E3313-E3322 ◽  
Author(s):  
Kevin J. P. Woods ◽  
Josh H. McDermott
Keyword(s):  
Time Varying ◽  
Cocktail Party ◽  
Sound Sources ◽  
Schema Learning ◽  
Cocktail Party Problem ◽  
Sound Structure ◽  
Natural Sound ◽  
Abstract Structure ◽  
Task Irrelevant ◽  
Acoustic Variation

The cocktail party problem requires listeners to infer individual sound sources from mixtures of sound. The problem can be solved only by leveraging regularities in natural sound sources, but little is known about how such regularities are internalized. We explored whether listeners learn source “schemas”—the abstract structure shared by different occurrences of the same type of sound source—and use them to infer sources from mixtures. We measured the ability of listeners to segregate mixtures of time-varying sources. In each experiment a subset of trials contained schema-based sources generated from a common template by transformations (transposition and time dilation) that introduced acoustic variation but preserved abstract structure. Across several tasks and classes of sound sources, schema-based sources consistently aided source separation, in some cases producing rapid improvements in performance over the first few exposures to a schema. Learning persisted across blocks that did not contain the learned schema, and listeners were able to learn and use multiple schemas simultaneously. No learning was evident when schema were presented in the task-irrelevant (i.e., distractor) source. However, learning from task-relevant stimuli showed signs of being implicit, in that listeners were no more likely to report that sources recurred in experiments containing schema-based sources than in control experiments containing no schema-based sources. The results implicate a mechanism for rapidly internalizing abstract sound structure, facilitating accurate perceptual organization of sound sources that recur in the environment.

Role of pinnae and head movements in localizing pure tones 1The authors would like to thank Mr. Remi Humbert for implementing the experiment in Turbo Pascal and for his further assistance.

1999 ◽  
Vol 58 (3) ◽  
pp. 170-179 ◽  
Author(s):  
Barbara S. Muller ◽  
Pierre Bovet
Keyword(s):  
Sound Source ◽  
Head Movement ◽  
Movement Analysis ◽  
Head Movements ◽  
Sound Sources ◽  
Localization Accuracy ◽  
Sound Effects ◽  
Posterior Quadrant ◽  
Localization Performance ◽  
Pure Tones

Twelve blindfolded subjects localized two different pure tones, randomly played by eight sound sources in the horizontal plane. Either subjects could get information supplied by their pinnae (external ear) and their head movements or not. We found that pinnae, as well as head movements, had a marked influence on auditory localization performance with this type of sound. Effects of pinnae and head movements seemed to be additive; the absence of one or the other factor provoked the same loss of localization accuracy and even much the same error pattern. Head movement analysis showed that subjects turn their face towards the emitting sound source, except for sources exactly in the front or exactly in the rear, which are identified by turning the head to both sides. The head movement amplitude increased smoothly as the sound source moved from the anterior to the posterior quadrant.

Sign in / Sign up

Export Citation Format

Share Document