• The Journal of the Acoustical Society of Korea
• /
• v.41 no.3
• /
• pp.326-334
• /
• 2022

In this paper, we propose speech enhancement algorithm as a pre-processing for robust speech recognition in noisy environments. Auxiliary-function-based Independent Vector Analysis (AuxIVA) is performed with weighted covariance matrix using time-varying variances with scaling factor from target masks representing time-frequency contributions of target speech. The mask estimates can be obtained using Neural Network (NN) pre-trained for speech extraction or diffuseness using Coherence-to-Diffuse power Ratio (CDR) to find the direct sounds component of a target speech. In addition, outputs for omni-directional noise are closely chained by sharing the time-varying variances similarly to independent subspace analysis or IVA. The speech extraction method based on AuxIVA is also performed in Independent Low-Rank Matrix Analysis (ILRMA) framework by extending the Non-negative Matrix Factorization (NMF) for noise outputs to Non-negative Tensor Factorization (NTF) to maintain the inter-channel dependency in noise output channels. Experimental results on the CHiME-4 datasets demonstrate the effectiveness of the presented algorithms.

• Journal of Environmental Policy
• /
• v.8 no.2
• /
• pp.83-118
• /
• 2009

Clean Development Mechanism(CDM) projects under the Kyoto Protocol have two objectives. One is to assist the Parties included in Annex I in achieving compliance with their quantified emission limitation and reduction commitments in cost-effective ways by allowing them to implement emission reduction projects in Non-Annex I countries and receive CERs, which will offset their reduction commitments. The other is to assist Parties not included in Annex I in achieving sustainable development and technology transfers through investments by Annex I countries. However, in reality, it is said that the former objective is achievable but the latter is not. In this light, this article suggests sustainability appraisal criteria applicable for Korea. Among various methodologies, we used the 'multi-attributes utility theory(MAUT)'; one of the 'multi-criteria analysis (MCA)' methodologies judged to be the most practical and relevant. Based on the guidelines of the MAUT methodology, we identified sustainability criteria that meet the guidelines. We took two tracks, the first to find the preferences of Korean experts, and the other to check foreign cases. In all, 37 preliminary criteria were suggested to Korean experts and each criterion was scored, from between 1 and 3, in terms of relevance, possibility of real improvement, easiness of data collection, and preferences. We combined foreign cases and the results of a survey conducted in Korea and selected 12 core criteria and 10 additional criteria. After that, all the criteria were converted into indicators. The indicators were applied to a CDM project for case study. We chose the "Sihwa Tidal Power Project", which is currently the biggest tidal power plant in the world. Twelve core indicators and 3 additional indicators were applied. In order to weight each indicator, the 'analytical hierarchy process (AHP)' was used. A total of 30 experts were asked to suggest weights and 21 answered. Among them, only 14 respondents were proven to meet the consistency ratio. We analyzed the 14 responses through Expert Choice and the CDM project was scored (+)53.082. In addition, sensitivity analysis was undertaken with the result of (+)44.667 to (+)65.522. As a result of this study, it was proven that this project would contribute to the sustainable development of Korea.