獨立成份分析法於可觸式聲波操控介面訊號分析之研究
No Thumbnail Available
Date
2013
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
本研究利用獨立成份分析法（ Independent Component Analysis，ICA）於可觸式聲
波操控介面上針對同時間不同觸碰事件所產生的混合訊號進行訊號分離。依據獨立成
份分析法的處理步驟，本研究提出三種分析操作程序來對可觸式聲波操控介面下麥克
風截取到的混合訊號實際進行測試，分別為 (一) 多音源輸入訊號直接進入ICA處理
(二) 單音源輸入訊號經頻率區隔分析後進行ICA處理 (三) 單音源輸入訊號經頻率區隔
及高次混合後進入ICA處理。研究結果顯示，透過頻率區隔在可觸式聲波操控介面下對
於非即時性的訊號分析是具有顯著的訊號分離效果。此結果也驗證了獨立成份分析法
可運用在可觸式聲波操控介面上作為混合訊號分離使用。
關鍵字：獨立成份分析法、可觸式聲波操控介面、聲源訊號分離
In this study, we tried to use independent component analysis (ICA) to separate the mixed signal caused by simultaneous touch events happened on a tangible acoustic interface. Being in accord with the processing steps of ICA, the procedures we proposed can be divided into (1) multiple audio inputs for ICA processing, (2) single audio input with frequency segmentation for ICA processing, (3) single audio input with frequency segmentation and higher-order mixed for ICA processing. Experimental results show that, using frequency segmentation for non-real time signal analysis on tangible acoustic interface has remarkable signal-separating effect. It also proves an idea that independent component analysis can be used on tangible acoustic interface for mixed-signal separation. Keywords : independent component analysis, tangible acoustic interface, audio signal separation
In this study, we tried to use independent component analysis (ICA) to separate the mixed signal caused by simultaneous touch events happened on a tangible acoustic interface. Being in accord with the processing steps of ICA, the procedures we proposed can be divided into (1) multiple audio inputs for ICA processing, (2) single audio input with frequency segmentation for ICA processing, (3) single audio input with frequency segmentation and higher-order mixed for ICA processing. Experimental results show that, using frequency segmentation for non-real time signal analysis on tangible acoustic interface has remarkable signal-separating effect. It also proves an idea that independent component analysis can be used on tangible acoustic interface for mixed-signal separation. Keywords : independent component analysis, tangible acoustic interface, audio signal separation
Description
Keywords
獨立成份分析法, 可觸式聲波操控介面, 聲源訊號分離, independent component analysis, tangible acoustic interface, audio signal separation