• The Journal of the Acoustical Society of Korea
• /
• v.40 no.5
• /
• pp.439-451
• /
• 2021

We propose a deep learning-based beamformer combined with spectral subtraction for continuous speech recognition operating in noisy environments. Conventional beamforming systems were mostly evaluated by using pre-segmented audio signals which were typically generated by mixing speech and noise continuously on a computer. However, since speech utterances are sparsely uttered along the time axis in real environments, conventional beamforming systems degrade in case when noise-only signals without speech are input. To alleviate this drawback, we combine online beamforming algorithm and spectral subtraction. We construct a Continuous Speech Enhancement (CSE) evaluation set to evaluate the online beamforming algorithm in noisy environments. The evaluation set is built by mixing sparsely-occurring speech utterances of the CHiME3 evaluation set and continuously-played CHiME3 background noise and background music of MUSDB. Using a Kaldi-based toolkit and Google web speech recognizer as a speech recognition back-end, we confirm that the proposed online beamforming algorithm with spectral subtraction shows better performance than the baseline online algorithm.

• The Journal of the Acoustical Society of Korea
• /
• v.18 no.1
• /
• pp.38-44
• /
• 1999

In this paper, a new spectral subtraction technique with two microphone inputs is proposed. In conventional spectral subtraction using a single microphone, the averaged noise spectrum is subtracted from the observed short-time input spectrum. This results in reduction of mean value of noise spectrum only, the component varying around the mean value remaining intact. In the method proposed in this paper, the short-time noise spectrum excluding the speech component is estimated by introducing the blocking matrix used in Griffiths-Jim-type adaptive beamformer with two microphone inputs, combined with the spectral compensation technique. A simulation was conducted to verify the effectiveness of the method.

• Journal of radiological science and technology
• /
• v.40 no.3
• /
• pp.443-451
• /
• 2017

Mammography is commonly used for screening early breast cancer. However, mammographic images, which depend on the physical properties of breast components, are limited to provide information about whether a lesion is malignant or benign. Although a dual-energy subtraction technique decomposes a certain material from a mixture, it increases radiation dose and degrades the accuracy of material decomposition. In this study, we simulated a breast phantom using attenuation characteristics, and we proposed a technique to enable the accurate material decomposition by applying weighting factors for the dual-energy mammography based on a photon-counting detector using a Monte Carlo simulation tool. We also evaluated the contrast and noise of simulated breast images for validating the proposed technique. As a result, the contrast for a malignant tumor in the dual-energy weighted subtraction technique was 0.98 and 1.06 times similar than those in the general mammography and dual-energy subtraction techniques, respectively. However the contrast between malignant and benign tumors dramatically increased 13.54 times due to the low contrast of a benign tumor. Therefore, the proposed technique can increase the material decomposition accuracy for malignant tumor and improve the diagnostic accuracy of mammography.