Spectral Subtraction Based on Non-extensive Statistics for Speech Recognition

スポンサーリンク

概要

• 論文の詳細を見る

• Spectral subtraction (SS) is an additive noise removal method which is derived in an extensive framework. In spectral subtraction, it is assumed that speech and noise spectra follow Gaussian distributions and are independent with each other. Hence, noisy speech also follows a Gaussian distribution. Spectral subtraction formula is obtained by maximizing the likelihood of noisy speech distribution with respect to its variance. However, it is well known that noisy speech observed in real situations often follows a heavy-tailed distribution, not a Gaussian distribution. In this paper, we introduce a q-Gaussian distribution in the non-extensive statistics to represent the distribution of noisy speech and derive a new spectral subtraction method based on it. We found that the q-Gaussian distribution fits the noisy speech distribution better than the Gaussian distribution does. Our speech recognition experiments using the Aurora-2 database showed that the proposed method, q-spectral subtraction (q-SS), outperformed the conventional SS method.

著者

• SHINODA Koichi

  Tokyo Institute of Technology

• Iwano Koji

  Tokyo City Univ. Yokohama‐shi Jpn

• IWANO Koji

  Tokyo City University

• PARDEDE Hilman

  Tokyo Institute of Technology

関連論文

• Acoustic Model Adaptation for Speech Recognition
• Acoustic Model Adaptation for Speech Recognition
• Committee-Based Active Learning for Speech Recognition
• Robust Gait-Based Person Identification against Walking Speed Variations
• Active Learning Using Phone-Error Distribution for Speech Modeling
• Spectral Subtraction Based on Non-extensive Statistics for Speech Recognition
• Active Learning Using Phone-Error Distribution for Speech Modeling

スポンサーリンク