臺灣博碩士論文加值系統

目前對於語音辨識方面的研究已有相當多的探討，伴隨著微電子技術的日新月異，超大型積體電路(Very Large Scale Integration﹔VLSI)得以快速發展﹔本文主要是以如何將語音辨識系統(The Voice Recognition System)實現為特殊應用積體電路(Application Specific Integrated Circuit﹔ASIC)為主體，所提出的新架構實現方法。在語音特徵值抽取與辨識法則裡使用的是線性預估法(Linear Predictor Coding；LPC)和類神經網路(Artificial Neural Network﹔ANN)。其中，類神經網路裡所使用的是漢明網路(Hamming Network)，全部共把整個語音識別的處理分為六個模組，以個別的方式逐一探討，並配合軟體模擬的方式加以驗證，期望以一個全新的方式來實現低面積、高辨識率的語音辨識晶片。然後，以TSMC 0.35 m 1P4M的製程技術，使用Galax! 0.35 m cell library來實現本文的架構。

Many topics had been studied in the field of voice recognition in recent year. With the dramatic revolution of the micro-electronic technology, Very Large Scale Integrated(VLSI) is developed very quickly. This paper focus on how to implement the voice recognition system as Application Specific Integrated Circuit(ASIC). The algorithm is Linear Predictor Coding(LPC) and Artificial Neural Network(ANN) in this paper. The Hamming Network is used in ANN. The whole processing of voice recognition is divided into six modules and is simulated with software package. The goal is to implement a chip with low area and high recognition rate in a whole new method. Then, the TSMC 0.35 m process technique is used to implement the design.

誌謝........................................................................ii
中文摘要...................................................................iii
Abstract....................................................................iv
目錄.........................................................................v
圖目錄.................................................................... vii
表目錄.......................................................................x
第一章 緒論..................................................................1
1.1前言....................................................................1
1.2研究動機................................................................1
1.3國內外有關本研究之情況..................................................2
第二章 語音辨識介紹..........................................................3
2.1結構流程................................................................3
2.2工作原理................................................................4
2.2.1外部模組........................................................4
LPF...........................................................4
Normalize.................................................... 4
A/D Converter.................................................4
2.2.2內部模組........................................................4
Sampling..................................................... 4
Pre-emphasis..................................................4
Windowing.....................................................5
頻譜分析......................................................6
類神經網路...................................................12
第三章 硬體架構.............................................................16
3.1晶片製作流程...........................................................16
3.2模組架構...............................................................17
3.2.1 Sampling、Pre-emphasis & Windowing.............................17
3.2.2 頻譜分析.......................................................20
3.2.3 類神經網路.....................................................23
3.2.4 總結...........................................................26
第四章 晶片實現.............................................................28
4.1 Function simulation...................................................28
4.2 Synthesis.............................................................30
4.3 Gate-Level simulation.................................................34
4.4 晶片佈局及包裝 .......................................................36
第五章 結論與未來展望.......................................................37
參考文獻....................................................................38
附錄........................................................................40

[1] 楊凰琳：“語音辨識晶片技術介紹，”電腦與通訊，47期，pp.58-69，1996。[2] Chris Rowden, ”Speech Processing,” McGraw-Hill, 1992.[3] 劉振源編譯，“類神經網路模型與語音辨識，”全華，1995。[4] W. Kinsner & C.Swanson, “Comparison of LPC Algorithms for Real-Time Speech Processing,” IEEE Engineering in Medicine & Biology Society 11th, Vol. 43, No. 3, pp. 759-771, Mar. 1995.[5] John R. Deller, JR. and John G. Proakis and John H. L. Hanse, “Discrete-Time Processing of Speech Signals,” Maxwell Macmillan International, 1993.[6] Martin T. Hagan, etc, “Neural Network Design,” PWS Publishing Co., 1995.[7] B.D.Ripley, “Pattern Recognition and Neural Networks,” Cambridge University Press, 1996. [8] Jeng-Feng Lan & Chung-Yu Wu, “The Multi-Chip Design of Analog CMOS Expandable Modified Hamming Neural Network with On-Chip Learning and Storage for Pattern Classification,” IEEE International Symposium on Circuits and Systems, June 9-12,1997, Hong Kong.[9] Behrooz Parhami, “Computer Arithmetic,” Oxford University Press, 2000.[10] Frederick Jelinek, “Tistical Methods for Speech Recognition,” Qualiton Imports, 1990.[11] Yoh-Han Pao, “Adaptive Pattern Recognition and Neural Networks,” Addison-Wesley, 1989. [12] C.Wheddon and R.Linggard, “Speech and Language Processing,” Van Nostrand Reinhold, 1990.[13] Lawrence Rabiner and Biing-Hwang Juang, “Fundamentals of Speech Recognition,” PTR Prentice Hall, 1993.[14] Nakamura, K. & Qiang Zhu & Maruoka, S. & Horiyama, T.& Kimura, S. & Watanabe, K., “Speech recognition chip for monosyllables,” IEEE Design Automation Conference, Proceedings of the ASP-DAC 2001, Asia and South Pacific.[15] Gordon E. Pelton, “Voice Processing,” McGraw-Hill, 1993.[16] J.N.Holmes, “Speech Synthesis and Recognition,” Van Nostrand Reinhold, 1988.[17] Furui, “Digital Speech Processing, Synthesis,and Recognition,” Marcel Dekker, 1989.[18] 張良羽，“語者辨識晶片之研製 (VLSI Design of Speaker Identification) ，”逢甲大學自動控制工程研究所碩士論文，1996。[19] 陳泰龍, “以高速模組為基礎之即時語音辨識系統單晶片設計(A VLSI Design of Real-Time Speech Recognition System Based on Fast Modules) ，”成功大學電機工程研究所碩士論文，2000。[20] 呂守仁, “個人語音識別系統-語音型樣之設計研究 (Individual Speech Recognition System —A Designing Research on Speech Pattern) ，”逢甲大學自動控制工程研究所碩士論文，1986。[21] 張嘉森：“使用語者分群模型的語音辨識方法，”電腦與通訊，86期，pp.9-16，2000。[22] 謝勝治編著，“圖控式程式語言-LabVIEW，”全華，1999。[23] 林傳生編著，“使用VHDL電路設計語言之數位電路設計，”儒林，1998。[24] 謝秀琴著，“數位語音信號基本原理，”全華，1996。[25] 張照煌等著：“非特定語者、中字彙中文語音辨識技術初探，”電腦與通訊，36期，pp.47-63，1995。[26] 陳東敏等著譯，“數位信號處理導論，”全華，1999。