Method of discriminating known and unknown environmental sounds using recurrent neural network

Yang Zhang, Tetsuya Ogata, Shun Nishide, Toru Takahashi, Hiroshi G. Okuno

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

This paper describes our method to classify nonspeech environmental sounds for robots working. In the real world, two main restrictions pertain in learning. First, robots have to learn using only a small amount of sounds in a limited time and space because of restrictions. Second, it has to detect unknown sounds to avoid false classification since it is virtually impossible to collect samples of all environmental sounds. Most of the previous methods require a huge number of samples of all target sounds, including noises, for training stochastic models such as the Gaussian mixture model. In contrast, we use a neurodynamical model to build a prediction and classification system. The neuro-dynamical system can be trained with a small amount of sounds and generalize others by inferring the sound generation dynamics. After training, a self-organized space is structured for the sound generation dynamics. The proposed system classify on the basis of the self-organized space. The prediction results of sounds are used for determining unknown sounds in our system. In this paper, we show the results of preliminary experiments on the proposed model's classification of known and unknown sound classes.

Original languageEnglish
Title of host publicationSCIS and ISIS 2010 - Joint 5th International Conference on Soft Computing and Intelligent Systems and 11th International Symposium on Advanced Intelligent Systems
Pages378-383
Number of pages6
Publication statusPublished - 2010
Externally publishedYes
EventJoint 5th International Conference on Soft Computing and Intelligent Systems and 11th International Symposium on Advanced Intelligent Systems, SCIS and ISIS 2010 - Okayama
Duration: 2010 Dec 82010 Dec 12

Other

OtherJoint 5th International Conference on Soft Computing and Intelligent Systems and 11th International Symposium on Advanced Intelligent Systems, SCIS and ISIS 2010
CityOkayama
Period10/12/810/12/12

Fingerprint

Recurrent neural networks
Acoustic waves
Robots
Stochastic models
Acoustic noise
Dynamical systems

ASJC Scopus subject areas

  • Artificial Intelligence
  • Information Systems

Cite this

Zhang, Y., Ogata, T., Nishide, S., Takahashi, T., & Okuno, H. G. (2010). Method of discriminating known and unknown environmental sounds using recurrent neural network. In SCIS and ISIS 2010 - Joint 5th International Conference on Soft Computing and Intelligent Systems and 11th International Symposium on Advanced Intelligent Systems (pp. 378-383)

Method of discriminating known and unknown environmental sounds using recurrent neural network. / Zhang, Yang; Ogata, Tetsuya; Nishide, Shun; Takahashi, Toru; Okuno, Hiroshi G.

SCIS and ISIS 2010 - Joint 5th International Conference on Soft Computing and Intelligent Systems and 11th International Symposium on Advanced Intelligent Systems. 2010. p. 378-383.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Zhang, Y, Ogata, T, Nishide, S, Takahashi, T & Okuno, HG 2010, Method of discriminating known and unknown environmental sounds using recurrent neural network. in SCIS and ISIS 2010 - Joint 5th International Conference on Soft Computing and Intelligent Systems and 11th International Symposium on Advanced Intelligent Systems. pp. 378-383, Joint 5th International Conference on Soft Computing and Intelligent Systems and 11th International Symposium on Advanced Intelligent Systems, SCIS and ISIS 2010, Okayama, 10/12/8.
Zhang Y, Ogata T, Nishide S, Takahashi T, Okuno HG. Method of discriminating known and unknown environmental sounds using recurrent neural network. In SCIS and ISIS 2010 - Joint 5th International Conference on Soft Computing and Intelligent Systems and 11th International Symposium on Advanced Intelligent Systems. 2010. p. 378-383
Zhang, Yang ; Ogata, Tetsuya ; Nishide, Shun ; Takahashi, Toru ; Okuno, Hiroshi G. / Method of discriminating known and unknown environmental sounds using recurrent neural network. SCIS and ISIS 2010 - Joint 5th International Conference on Soft Computing and Intelligent Systems and 11th International Symposium on Advanced Intelligent Systems. 2010. pp. 378-383
@inproceedings{b52b9cb125f743509cb396c811163946,
title = "Method of discriminating known and unknown environmental sounds using recurrent neural network",
abstract = "This paper describes our method to classify nonspeech environmental sounds for robots working. In the real world, two main restrictions pertain in learning. First, robots have to learn using only a small amount of sounds in a limited time and space because of restrictions. Second, it has to detect unknown sounds to avoid false classification since it is virtually impossible to collect samples of all environmental sounds. Most of the previous methods require a huge number of samples of all target sounds, including noises, for training stochastic models such as the Gaussian mixture model. In contrast, we use a neurodynamical model to build a prediction and classification system. The neuro-dynamical system can be trained with a small amount of sounds and generalize others by inferring the sound generation dynamics. After training, a self-organized space is structured for the sound generation dynamics. The proposed system classify on the basis of the self-organized space. The prediction results of sounds are used for determining unknown sounds in our system. In this paper, we show the results of preliminary experiments on the proposed model's classification of known and unknown sound classes.",
author = "Yang Zhang and Tetsuya Ogata and Shun Nishide and Toru Takahashi and Okuno, {Hiroshi G.}",
year = "2010",
language = "English",
pages = "378--383",
booktitle = "SCIS and ISIS 2010 - Joint 5th International Conference on Soft Computing and Intelligent Systems and 11th International Symposium on Advanced Intelligent Systems",

}

TY - GEN

T1 - Method of discriminating known and unknown environmental sounds using recurrent neural network

AU - Zhang, Yang

AU - Ogata, Tetsuya

AU - Nishide, Shun

AU - Takahashi, Toru

AU - Okuno, Hiroshi G.

PY - 2010

Y1 - 2010

N2 - This paper describes our method to classify nonspeech environmental sounds for robots working. In the real world, two main restrictions pertain in learning. First, robots have to learn using only a small amount of sounds in a limited time and space because of restrictions. Second, it has to detect unknown sounds to avoid false classification since it is virtually impossible to collect samples of all environmental sounds. Most of the previous methods require a huge number of samples of all target sounds, including noises, for training stochastic models such as the Gaussian mixture model. In contrast, we use a neurodynamical model to build a prediction and classification system. The neuro-dynamical system can be trained with a small amount of sounds and generalize others by inferring the sound generation dynamics. After training, a self-organized space is structured for the sound generation dynamics. The proposed system classify on the basis of the self-organized space. The prediction results of sounds are used for determining unknown sounds in our system. In this paper, we show the results of preliminary experiments on the proposed model's classification of known and unknown sound classes.

AB - This paper describes our method to classify nonspeech environmental sounds for robots working. In the real world, two main restrictions pertain in learning. First, robots have to learn using only a small amount of sounds in a limited time and space because of restrictions. Second, it has to detect unknown sounds to avoid false classification since it is virtually impossible to collect samples of all environmental sounds. Most of the previous methods require a huge number of samples of all target sounds, including noises, for training stochastic models such as the Gaussian mixture model. In contrast, we use a neurodynamical model to build a prediction and classification system. The neuro-dynamical system can be trained with a small amount of sounds and generalize others by inferring the sound generation dynamics. After training, a self-organized space is structured for the sound generation dynamics. The proposed system classify on the basis of the self-organized space. The prediction results of sounds are used for determining unknown sounds in our system. In this paper, we show the results of preliminary experiments on the proposed model's classification of known and unknown sound classes.

UR - http://www.scopus.com/inward/record.url?scp=84866693937&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84866693937&partnerID=8YFLogxK

M3 - Conference contribution

SP - 378

EP - 383

BT - SCIS and ISIS 2010 - Joint 5th International Conference on Soft Computing and Intelligent Systems and 11th International Symposium on Advanced Intelligent Systems

ER -