Automatic estimation of reverberation time with robot speech to improve ICA-based robot audition

Ryu Takeda, Kazuhiro Nakadai, Toru Takahashi, Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno

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

1 Citation (Scopus)

Abstract

This paper presents an ICA-based robot audition system which estimates the reverberation time of the environment automatically by using the robot's own speech. The system is based on multi-channel semi-blind independent component analysis (MCSB-ICA), a source separation method using a microphone array that can separate user and robot speech under reverberant environments. Perception of the reverberation time (RT) is critical, because an inappropriate RT degrades separation performance and increases processing time. Unlike most previous methods that assume the RT is given in advance, our method estimates an RT by using the echo's intensity of the robot's own speech. It has three steps: speaks a sentence in a new environment, calculates the relative powers of the echoes, and estimates the RT using linear regression of them. Experimental results show that this method sets an appropriate RT for MCSB-ICA for real-world environments and that word correctness is improved by up to 6 points and processing time is reduced by up to 60%.

Original languageEnglish
Title of host publication9th IEEE-RAS International Conference on Humanoid Robots, HUMANOIDS09
Pages250-255
Number of pages6
DOIs
Publication statusPublished - 2009
Externally publishedYes
Event9th IEEE-RAS International Conference on Humanoid Robots, HUMANOIDS09 - Paris
Duration: 2009 Dec 72009 Dec 10

Other

Other9th IEEE-RAS International Conference on Humanoid Robots, HUMANOIDS09
CityParis
Period09/12/709/12/10

Fingerprint

Reverberation
Independent component analysis
Audition
Robots
Source separation
Microphones
Processing
Linear regression

ASJC Scopus subject areas

  • Computer Science(all)

Cite this

Takeda, R., Nakadai, K., Takahashi, T., Komatani, K., Ogata, T., & Okuno, H. G. (2009). Automatic estimation of reverberation time with robot speech to improve ICA-based robot audition. In 9th IEEE-RAS International Conference on Humanoid Robots, HUMANOIDS09 (pp. 250-255). [5379572] https://doi.org/10.1109/ICHR.2009.5379572

Automatic estimation of reverberation time with robot speech to improve ICA-based robot audition. / Takeda, Ryu; Nakadai, Kazuhiro; Takahashi, Toru; Komatani, Kazunori; Ogata, Tetsuya; Okuno, Hiroshi G.

9th IEEE-RAS International Conference on Humanoid Robots, HUMANOIDS09. 2009. p. 250-255 5379572.

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

Takeda, R, Nakadai, K, Takahashi, T, Komatani, K, Ogata, T & Okuno, HG 2009, Automatic estimation of reverberation time with robot speech to improve ICA-based robot audition. in 9th IEEE-RAS International Conference on Humanoid Robots, HUMANOIDS09., 5379572, pp. 250-255, 9th IEEE-RAS International Conference on Humanoid Robots, HUMANOIDS09, Paris, 09/12/7. https://doi.org/10.1109/ICHR.2009.5379572
Takeda R, Nakadai K, Takahashi T, Komatani K, Ogata T, Okuno HG. Automatic estimation of reverberation time with robot speech to improve ICA-based robot audition. In 9th IEEE-RAS International Conference on Humanoid Robots, HUMANOIDS09. 2009. p. 250-255. 5379572 https://doi.org/10.1109/ICHR.2009.5379572
Takeda, Ryu ; Nakadai, Kazuhiro ; Takahashi, Toru ; Komatani, Kazunori ; Ogata, Tetsuya ; Okuno, Hiroshi G. / Automatic estimation of reverberation time with robot speech to improve ICA-based robot audition. 9th IEEE-RAS International Conference on Humanoid Robots, HUMANOIDS09. 2009. pp. 250-255
@inproceedings{8b61da8361674b30bebc39592fc43d1f,
title = "Automatic estimation of reverberation time with robot speech to improve ICA-based robot audition",
abstract = "This paper presents an ICA-based robot audition system which estimates the reverberation time of the environment automatically by using the robot's own speech. The system is based on multi-channel semi-blind independent component analysis (MCSB-ICA), a source separation method using a microphone array that can separate user and robot speech under reverberant environments. Perception of the reverberation time (RT) is critical, because an inappropriate RT degrades separation performance and increases processing time. Unlike most previous methods that assume the RT is given in advance, our method estimates an RT by using the echo's intensity of the robot's own speech. It has three steps: speaks a sentence in a new environment, calculates the relative powers of the echoes, and estimates the RT using linear regression of them. Experimental results show that this method sets an appropriate RT for MCSB-ICA for real-world environments and that word correctness is improved by up to 6 points and processing time is reduced by up to 60{\%}.",
author = "Ryu Takeda and Kazuhiro Nakadai and Toru Takahashi and Kazunori Komatani and Tetsuya Ogata and Okuno, {Hiroshi G.}",
year = "2009",
doi = "10.1109/ICHR.2009.5379572",
language = "English",
isbn = "9781424445882",
pages = "250--255",
booktitle = "9th IEEE-RAS International Conference on Humanoid Robots, HUMANOIDS09",

}

TY - GEN

T1 - Automatic estimation of reverberation time with robot speech to improve ICA-based robot audition

AU - Takeda, Ryu

AU - Nakadai, Kazuhiro

AU - Takahashi, Toru

AU - Komatani, Kazunori

AU - Ogata, Tetsuya

AU - Okuno, Hiroshi G.

PY - 2009

Y1 - 2009

N2 - This paper presents an ICA-based robot audition system which estimates the reverberation time of the environment automatically by using the robot's own speech. The system is based on multi-channel semi-blind independent component analysis (MCSB-ICA), a source separation method using a microphone array that can separate user and robot speech under reverberant environments. Perception of the reverberation time (RT) is critical, because an inappropriate RT degrades separation performance and increases processing time. Unlike most previous methods that assume the RT is given in advance, our method estimates an RT by using the echo's intensity of the robot's own speech. It has three steps: speaks a sentence in a new environment, calculates the relative powers of the echoes, and estimates the RT using linear regression of them. Experimental results show that this method sets an appropriate RT for MCSB-ICA for real-world environments and that word correctness is improved by up to 6 points and processing time is reduced by up to 60%.

AB - This paper presents an ICA-based robot audition system which estimates the reverberation time of the environment automatically by using the robot's own speech. The system is based on multi-channel semi-blind independent component analysis (MCSB-ICA), a source separation method using a microphone array that can separate user and robot speech under reverberant environments. Perception of the reverberation time (RT) is critical, because an inappropriate RT degrades separation performance and increases processing time. Unlike most previous methods that assume the RT is given in advance, our method estimates an RT by using the echo's intensity of the robot's own speech. It has three steps: speaks a sentence in a new environment, calculates the relative powers of the echoes, and estimates the RT using linear regression of them. Experimental results show that this method sets an appropriate RT for MCSB-ICA for real-world environments and that word correctness is improved by up to 6 points and processing time is reduced by up to 60%.

UR - http://www.scopus.com/inward/record.url?scp=77950583451&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77950583451&partnerID=8YFLogxK

U2 - 10.1109/ICHR.2009.5379572

DO - 10.1109/ICHR.2009.5379572

M3 - Conference contribution

AN - SCOPUS:77950583451

SN - 9781424445882

SP - 250

EP - 255

BT - 9th IEEE-RAS International Conference on Humanoid Robots, HUMANOIDS09

ER -