USING DYNAMIC CONDITIONAL RANDOM FIELD ON SINGLE-MICROPHONE SPEECH SEPARATION
Refereed conference paper presented and published in conference proceedings

香港中文大學研究人員

全文

引用次數

其它資訊
摘要The use of dynamic conditional random field (DCRF) for model-based single-microphone speech separation is investigated. The speech sources are represented by acoustic state sequences from speaker-dependent acoustic models. The posterior probabilities of the source acoustic states given a speech mixture are inferred with a maximum entropy probability distribution which is represented by DCRF. The posterior probabilities are needed for minimum mean-square error estimation of the speech sources. Loopy belief propagation is applied for the inference. Averaged stochastic gradient descent and limited-memory BFGS are compared for parameter estimation. With the log-magnitude spectrum of the speech mixture as input observation, the proposed method achieves better separation performance in terms of Blind Source Separation Metrics (SDR, SAR, SIR) and PESQ than a factorial hidden Markov model baseline system in our experiments.
著者Yeung YT, Lee T, Leung CC
會議名稱IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP)
會議開始日26.05.2013
會議完結日31.05.2013
會議地點Vancouver
會議國家/地區加拿大
詳細描述organized by IEEE,
出版年份2013
月份1
日期1
出版社IEEE
頁次146 - 150
電子國際標準書號978-1-4799-0356-6
國際標準期刊號1520-6149
語言英式英語
關鍵詞dynamic conditional random field; single-microphone speech separation
Web of Science 學科類別Acoustics; Engineering; Engineering, Electrical & Electronic

上次更新時間 2021-26-02 於 00:18