國立高雄大學圖資館 |

語系: 繁體中文

說明(常見問題)

圖資館首頁

登入

回首頁

切換: 標籤 | MARC模式 | ISBD

具有強健性語音辨識的無線語音控制系統研製 = Robust Speech...

國立高雄大學資訊工程學系碩士班

具有強健性語音辨識的無線語音控制系統研製 = Robust Speech Recognition on Wireless Speech Control System

紀錄類型:	書目-語言資料,印刷品 : 單行本
並列題名:	Robust Speech Recognition on Wireless Speech Control System
作者:	蔡宜亨,
其他團體作者:	國立高雄大學
出版地:	[高雄市]
出版者:	撰者;
出版年:	民100
面頁冊數:	91葉圖，表格 : 30公分;
標題:	語音辨識
標題:	Speech Recognition
電子資源:	http://handle.ncl.edu.tw/11296/ndltd/18985882401038093546
附註:	106年10月31日公開
附註:	參考書目：葉80-82
摘要註:	本論文主要是以提升語音的辨識率與語音的抗雜訊能力。在提升語音辨識率方面，是以模糊向量量化(Fuzzy Vector Quantization, FVQ)來建立離散隱藏式馬可夫模型(DHMM)並用於語音模型的訓練，同樣以模糊向量量化(Fuzzy Vector Quantization, FVQ)用於輸入語音的辨識。而語音抗雜訊方面，是以經驗模態分解法(Empirical Mode Decomposition, EMD) 將含雜訊的語音訊號分解成多組本質模態函式(Intrinsic Mode Function, IMF)，並以實數型基因演算法找出最佳IMF組合參數，再將分離出之IMF依組合參數還原成語音。結合FVQ和EMD之特點，將使得具有雜訊的語音辨識率提升很多。另外，我們在FPGA平台使用語音關鍵詞方式來控制清潔機器人。本論文透過無線模組將FPGA的辨識結果傳送至清潔用機器人上的AT89S51處理器，來控制清潔機器人的動作，讓清潔機器人可以在吵雜環境中執行正確的動作。 This thesis is to enhance and improve speech recognition rate in noised environment. We use fuzzy vector quantization (FVQ) to improve the modeling of Discrete Hidden Markov Model (DHMM) and then to improve the speech recognition rate. The Empirical Mode Decomposition (EMD) is used to discompose some noised speech signals into several Intrinsic Mode Functions (IMF). These IMFs will be combined to recover the original speech by multiplying their corresponding weights which were trained by Genetic Algorithms (GA). After applying Empirical Mode Decomposition (EMD), we obtain a cleaner speech for recognition. Combining EMD and FVQ, we can improve the noised speech recognition rate. Besides, we utilize speech keywords to control the clean robot by using FPGA platform. So, the clean robot can perform the correct action in the noised environment. In this research, a wireless module is used to transmit the control comments form FPGA to the AT89S51 processer embedded in the clean robot to control the behavior of the robot.

具有強健性語音辨識的無線語音控制系統研製 = Robust Speech Recognition on Wireless Speech Control System
蔡, 宜亨

具有強健性語音辨識的無線語音控制系統研製 = Robust Speech Recognition on Wireless Speech Control System / 蔡宜亨撰 - [高雄市] : 撰者, 民100. - 91葉 ; 圖，表格 ; 30公分.
106年10月31日公開參考書目：葉80-82.
語音辨識Speech Recognition

具有強健性語音辨識的無線語音控制系統研製 = Robust Speech Recognition on Wireless Speech Control System
LDR:03322nam0a2200289 450 001 300435
005 20171103092536.0
009 300435
010 0 $b 精裝
010 0 $b 平裝
100 $a 20111013d2011 k y0chiy50 e
101 1 $a chi $d chi $d eng
102 $a tw
105 $a ak am 000yy
200 1 $a 具有強健性語音辨識的無線語音控制系統研製 $d Robust Speech Recognition on Wireless Speech Control System $z eng $f 蔡宜亨撰
210 $a [高雄市] $c 撰者 $d 民100
215 0 $a 91葉 $c 圖，表格 $d 30公分
300 $a 106年10月31日公開
300 $a 參考書目：葉80-82
314 $a 指導教授：潘欣泰博士
328 $a 碩士論文--國立高雄大學資訊工程學系碩士班
330 $a 本論文主要是以提升語音的辨識率與語音的抗雜訊能力。在提升語音辨識率方面，是以模糊向量量化(Fuzzy Vector Quantization, FVQ)來建立離散隱藏式馬可夫模型(DHMM)並用於語音模型的訓練，同樣以模糊向量量化(Fuzzy Vector Quantization, FVQ)用於輸入語音的辨識。而語音抗雜訊方面，是以經驗模態分解法(Empirical Mode Decomposition, EMD) 將含雜訊的語音訊號分解成多組本質模態函式(Intrinsic Mode Function, IMF)，並以實數型基因演算法找出最佳IMF組合參數，再將分離出之IMF依組合參數還原成語音。結合FVQ和EMD之特點，將使得具有雜訊的語音辨識率提升很多。另外，我們在FPGA平台使用語音關鍵詞方式來控制清潔機器人。本論文透過無線模組將FPGA的辨識結果傳送至清潔用機器人上的AT89S51處理器，來控制清潔機器人的動作，讓清潔機器人可以在吵雜環境中執行正確的動作。 This thesis is to enhance and improve speech recognition rate in noised environment. We use fuzzy vector quantization (FVQ) to improve the modeling of Discrete Hidden Markov Model (DHMM) and then to improve the speech recognition rate. The Empirical Mode Decomposition (EMD) is used to discompose some noised speech signals into several Intrinsic Mode Functions (IMF). These IMFs will be combined to recover the original speech by multiplying their corresponding weights which were trained by Genetic Algorithms (GA). After applying Empirical Mode Decomposition (EMD), we obtain a cleaner speech for recognition. Combining EMD and FVQ, we can improve the noised speech recognition rate. Besides, we utilize speech keywords to control the clean robot by using FPGA platform. So, the clean robot can perform the correct action in the noised environment. In this research, a wireless module is used to transmit the control comments form FPGA to the AT89S51 processer embedded in the clean robot to control the behavior of the robot.
510 1 $a Robust Speech Recognition on Wireless Speech Control System $z eng
610 0 $a 語音辨識 $a 經驗模態分解法 $a 模糊向量量化 $a 離散隱藏式馬可夫模型 $a 基因演算法 $a 場域可程式閘極陣列
610 1 $a Speech Recognition $a Empirical Mode Decomposition $a Fuzzy Vector Quantization $a Discrete Hidden Markov Model $a Genetic Algorithms $a Field Programmable Gate Array
681 $a 008M/0019 $b 464103 4430 $v 2007年版
700 1 $a 蔡 $b 宜亨 $4 撰 $3 518827
712 0 2 $a 國立高雄大學 $b 資訊工程學系碩士班 $3 353878
801 0 $a tw $b NUK $c 20111027 $g CCR
856 7 $z 電子資源 $2 http $u http://handle.ncl.edu.tw/11296/ndltd/18985882401038093546