VQ-based Approach to Single-Channel Audio Separation for Music and Speech Mixtures

سال انتشار: 1388
نوع سند: مقاله ژورنالی
زبان: انگلیسی
مشاهده: 99

فایل این مقاله در 10 صفحه با فرمت PDF قابل دریافت می باشد

استخراج به نرم افزارهای پژوهشی:

لینک ثابت به این مقاله:

شناسه ملی سند علمی:

JR_ITRC-2-1_001

تاریخ نمایه سازی: 23 فروردین 1401

چکیده مقاله:

In this paper, we propose a low-complexity model-based single-channel audio separation approach. The proposed method presents three certain advantages over previous methods: I) replacing commonly used linear masks like Wiener filtering by a proposed non-linear one, we show that it is possible to lower the crosstalk of the interfering source often occurring in a mask-based method while recovering the underlying signals from the observed mixture. Using nonlinear masks establishes a tradeoff between acceptable level of interference and low speech distortion, ۲) as a post-processing stage, we use phase synchronization technique to enhance the perceptual quality of the re synthesized signals, and ۳) the proposed method is based on vector quantization {VQ) codebooks. Hence, the complexity is lower than previous GMM-based methods. Through extensive experiments, it is demonstrated that the proposed method can achieve a lower signal-to-distortion ratio (SDR). According to our listening experiments and according to the Mean Opinion Score (MOS) results, it is confirmed that the proposed method is able to recover separated outputs with a higher perceived signal quality.

نویسندگان

Pejman Mowlaee

Electrical Engineering Department Amirkabir University of Technology Tehran, Iran

Abolghasem Sayadiyan

Electrical Engineering Department Amirkabir University of Technology Tehran, Iran

Hamid Sheikhzadeh Nadjar

Electrical Engineering Department Amirkabir University of Technology Tehran, Iran