Inventory based speech enhancement for speaker dedicated speech communication systems

An iterative longest matching segment approach to speech enhancement with additive noise and channel distortion

“…6. For our test data, the 23 histogram became largely stable after four iterations. In our experiments, we stopped the iterations 24 for each test utterance if successive iterations produced longest matched segment estimates with 25 similar average lengths.…”

“…As a result, they require specific knowledge about the noise 23 for noise removal and hence speech recovery. The typical constraint or prior for the underlying 24 speech is the probability distribution of the speech short-time discrete Fourier transform (DFT) 25 coefficients or spectral amplitudes (e.g., Cohen, 2005;Ephraim and Malah, 1984;Lotter and Vary, 26 2005; Martin, 2002;Martin and Breithaupt, 2003).…”

“…We demonstrate the improved performance for the 22 new approach through experiments for speech recognition and speech enhancement. 23 The remainder of this paper is organized as follows. Section 2 outlines the assumptions made 24 in this research for modeling noisy speech, and the key idea of the proposed approach for speech 25 estimation.…”

“…We 22 use a simplified example to illustrate our idea. Consider the power spectral density (PSD) as the 23 statistics of a signal in the linear-spectral domain. Suppose Fig.…”

“…For speech enhancement, while Wiener filtering based on the matched corpus speech frame, as 23 described in (Ming et al, 2011), can be used to suppress the additive noise, it is not effective for 24 recovering speech from channel distortion. Therefore, in our speech enhancement experiments with 25 the Aurora 4 database, we reconstruct the waveform for each underlying speech frame by using 26 the magnitude spectrum of the matched corpus speech frame.…”

An iterative longest matching segment approach to speech enhancement with additive noise and channel distortion

“…6. For our test data, the 23 histogram became largely stable after four iterations. In our experiments, we stopped the iterations 24 for each test utterance if successive iterations produced longest matched segment estimates with 25 similar average lengths.…”

“…As a result, they require specific knowledge about the noise 23 for noise removal and hence speech recovery. The typical constraint or prior for the underlying 24 speech is the probability distribution of the speech short-time discrete Fourier transform (DFT) 25 coefficients or spectral amplitudes (e.g., Cohen, 2005;Ephraim and Malah, 1984;Lotter and Vary, 26 2005; Martin, 2002;Martin and Breithaupt, 2003).…”

“…We demonstrate the improved performance for the 22 new approach through experiments for speech recognition and speech enhancement. 23 The remainder of this paper is organized as follows. Section 2 outlines the assumptions made 24 in this research for modeling noisy speech, and the key idea of the proposed approach for speech 25 estimation.…”

“…We 22 use a simplified example to illustrate our idea. Consider the power spectral density (PSD) as the 23 statistics of a signal in the linear-spectral domain. Suppose Fig.…”

“…For speech enhancement, while Wiener filtering based on the matched corpus speech frame, as 23 described in (Ming et al, 2011), can be used to suppress the additive noise, it is not effective for 24 recovering speech from channel distortion. Therefore, in our speech enhancement experiments with 25 the Aurora 4 database, we reconstruct the waveform for each underlying speech frame by using 26 the magnitude spectrum of the matched corpus speech frame.…”

An iterative longest matching segment approach to speech enhancement with additive noise and channel distortion

“…6. For our test data, the 23 histogram became largely stable after four iterations. In our experiments, we stopped the iterations 24 for each test utterance if successive iterations produced longest matched segment estimates with 25 similar average lengths.…”

“…As a result, they require specific knowledge about the noise 23 for noise removal and hence speech recovery. The typical constraint or prior for the underlying 24 speech is the probability distribution of the speech short-time discrete Fourier transform (DFT) 25 coefficients or spectral amplitudes (e.g., Cohen, 2005;Ephraim and Malah, 1984;Lotter and Vary, 26 2005; Martin, 2002;Martin and Breithaupt, 2003).…”

“…We demonstrate the improved performance for the 22 new approach through experiments for speech recognition and speech enhancement. 23 The remainder of this paper is organized as follows. Section 2 outlines the assumptions made 24 in this research for modeling noisy speech, and the key idea of the proposed approach for speech 25 estimation.…”

“…We 22 use a simplified example to illustrate our idea. Consider the power spectral density (PSD) as the 23 statistics of a signal in the linear-spectral domain. Suppose Fig.…”

“…For speech enhancement, while Wiener filtering based on the matched corpus speech frame, as 23 described in (Ming et al, 2011), can be used to suppress the additive noise, it is not effective for 24 recovering speech from channel distortion. Therefore, in our speech enhancement experiments with 25 the Aurora 4 database, we reconstruct the waveform for each underlying speech frame by using 26 the magnitude spectrum of the matched corpus speech frame.…”

An iterative longest matching segment approach to speech enhancement with additive noise and channel distortion

“…6. For our test data, the 23 histogram became largely stable after four iterations. In our experiments, we stopped the iterations 24 for each test utterance if successive iterations produced longest matched segment estimates with 25 similar average lengths.…”

“…As a result, they require specific knowledge about the noise 23 for noise removal and hence speech recovery. The typical constraint or prior for the underlying 24 speech is the probability distribution of the speech short-time discrete Fourier transform (DFT) 25 coefficients or spectral amplitudes (e.g., Cohen, 2005;Ephraim and Malah, 1984;Lotter and Vary, 26 2005; Martin, 2002;Martin and Breithaupt, 2003).…”

“…We demonstrate the improved performance for the 22 new approach through experiments for speech recognition and speech enhancement. 23 The remainder of this paper is organized as follows. Section 2 outlines the assumptions made 24 in this research for modeling noisy speech, and the key idea of the proposed approach for speech 25 estimation.…”

“…We 22 use a simplified example to illustrate our idea. Consider the power spectral density (PSD) as the 23 statistics of a signal in the linear-spectral domain. Suppose Fig.…”

“…For speech enhancement, while Wiener filtering based on the matched corpus speech frame, as 23 described in (Ming et al, 2011), can be used to suppress the additive noise, it is not effective for 24 recovering speech from channel distortion. Therefore, in our speech enhancement experiments with 25 the Aurora 4 database, we reconstruct the waveform for each underlying speech frame by using 26 the magnitude spectrum of the matched corpus speech frame.…”