INTERSPEECH 2008
9th Annual Conference of the International Speech Communication Association

Brisbane, Australia
September 22-26, 2008

CENSREC-4: Development of Evaluation Framework for Distant-Talking Speech Recognition Under Reverberant Environments

Masato Nakayama (1), Takanobu Nishiura (1), Yuki Denda (1), Norihide Kitaoka (2), Kazumasa Yamamoto (3), Takeshi Yamada (4), Satoru Tsuge (5), Chiyomi Miyajima (2), Masakiyo Fujimoto (6), Tetsuya Takiguchi (7), Satoshi Tamura (8), Tetsuji Ogawa (9), Shigeki Matsuda (10), Shingo Kuroiwa (11), Kazuya Takeda (2), Satoshi Nakamura (10)

(1) Ritsumeikan University, Japan; (2) Nagoya University, Japan; (3) Toyohashi University of Technology, Japan; (4) University of Tsukuba, Japan; (5) University of Tokushima, Japan; (6) NTT Corporation, Japan; (7) Kobe University, Japan; (8) Gifu University, Japan; (9) Waseda University, Japan; (10) ATR-SLC, Japan; (11) Chiba University, Japan

In this paper, we newly introduce a collection of databases and evaluation tools called CENSREC-4, which is an evaluation framework for distant-talking speech under hands-free conditions. Distant-talking speech recognition is crucial for a hands-free speech interface. Therefore, we measured room impulse responses to investigate reverberant speech recognition in various environments. The data contained in CENSREC-4 are connected digit utterances, as in CENSREC-1. Two subsets are included in the data: basic data sets and extra data sets. The basic data sets are used for the evaluation environment for the room impulse response-convolved speech data. The extra data sets consist of simulated and recorded data. An evaluation framework is only provided for the basic data sets as evaluation tools. The results of evaluation experiments proved that CENSREC-4 is an effective database for evaluating the new dereverberation method because the traditional dereverberation process had difficulty sufficiently improving the recognition performance.

Full Paper

Bibliographic reference.  Nakayama, Masato / Nishiura, Takanobu / Denda, Yuki / Kitaoka, Norihide / Yamamoto, Kazumasa / Yamada, Takeshi / Tsuge, Satoru / Miyajima, Chiyomi / Fujimoto, Masakiyo / Takiguchi, Tetsuya / Tamura, Satoshi / Ogawa, Tetsuji / Matsuda, Shigeki / Kuroiwa, Shingo / Takeda, Kazuya / Nakamura, Satoshi (2008): "CENSREC-4: development of evaluation framework for distant-talking speech recognition under reverberant environments", In INTERSPEECH-2008, 968-971.