臺灣博碩士論文加值系統

本篇論文主要目標是探討在語音編碼在通訊傳輸中，當某些語音框架編碼遺失，如何重整語音訊號，以達到最佳效果。論文選擇碼簿激發線性預測編碼(CELP)為主要研究對象，使用FS1016 4.8kbps 低碼率高品質CELP語音編碼為遺失語音編碼框架後重整語音訊號分析之語音編碼方式，在語音訊號重整方式方面，考量速度等問題，將以遺失語音框架編碼重建做為主要語音訊號重整的方式。因此遺失語音框架編碼重建的工作主要分為無任何遺失語音框架編碼資訊及保有部分遺失語音框架編碼資訊兩類，用實驗分析在各種不同狀況下重建語音訊號的方法，並探討CELP語音編碼各部分參數在遺失語音框架編碼時訊號重整的重要性，然後藉著保留這些較重要的參數，以便在語音框架編碼遺失的時候來重建出較佳的語音品質。研究結果顯示，音高增益(pitch gain)係數在CELP語音編碼中，語音框架編碼遺失後語音訊號重整，扮演最重要因素。適度保留失語音編碼框架中的音高增益(pitch gain)及LSF係數，即使在高音框編碼遺失率下(30%)，還可以還原回可接受的語音訊號。最後將研究結果，撰寫成聲音管理員中語音編碼器(ACM codec)，並選用Netmeeting來做網路即時語音傳輸測試，並將語音編碼器程式轉移至DSP晶片TMS320C54X上。

This thesis is mainly to explore the best ways to do the voice reconstruction when the transmitted speech signal in speech coding lost some frame’s codewords. In this thesis, we focus on speech coding method based on code excited linear prediction(CELP), and use the high quality low bit-rate speech vocoder -- FS1016 4.8 kbps CELP as the speech coding methods to compare the performance in the voice reconstruction under loss of frame’s codewords Considering the factors for speeding and simplicity, we use the lost frame’s codewords reconstruction as the major way for voice reconstruction under the loss of frame’s codewords. The methods for lost frame’s codewords reconstruction can be classified into two groups. One is the lost frame’s codewords reconstruction without any information of the lost frame’s codewords, the other is the lost frame’s codewords reconstruction with partial information of the lost frame’s codewords. We do experiments to investigate he best ways to do the voice reconstruction in varied situations. And we also study the importance of the parameters in a CELP frame’s codeword in voice reconstruction and improve the voice reconstruction’s quality by keeping the most important partial information in the lost frame’s codewords. Our results show that the most important parameter in the voice reconstruction due to the lost frame’s codewords is the pitch gain. By adequately keeping the parameters of pitch gain and LSF coefficients in the lost frames, we can reconstruct acceptable speech signal, evenly at high frame lost rate(30%). Finally, we implement our results as an ACM codec, and use Netmeeting to do the real-time voice transmission in the internet. We also port our programs to TI DSP chip TMS320C54x system.

目 次
摘要（中文）
摘要（英文）
誌謝
目次
圖目錄
表目錄
第一章 緒論
1.1 簡介
1.1.1 語音訊號的產生
1.1.2 數位發音模型
1.1.3 典型語音訊號
1.2 研究方法
1.3 各章提要
第二章 語音編碼原理
2.1 簡介
2.1.1 波形編碼
2.1.2 語音模型編碼
2.1.3 混和編碼
2.2 語音編碼的標準
2.3 語音品質的評估
2.4 FS 1016 4.8 kbps CELP簡介
2.5 LPC 分析
2.5.1 語音信號的線性預估模型化
2.5.2 LPC參數的求解
2.5.3 LPC參數的量化
2.5.3.1 LSF係數
2.5.3.2 LPC參數轉換成LSF係數
2.5.3.3 LSF係數的內插
2.6 感官加權濾波
2.7 音高預測
2.8 雜訊碼簿搜尋
2.9 後濾波處理器
第三章 語音資料庫
3.1 簡介
3.2 男性語者語音訊號
3.3 女性語者語音訊號
第四章 CELP語音編碼框架遺失之訊號重整
4.1 簡介
4.2 不保留遺失音框編碼參數之訊號重整
4.2.1 研究動機
4.2.2 實驗結果
4.2.3 結論
4.3 保留遺失音框編碼LSF係數之訊號重整
4.3.1 研究動機
4.3.2 實驗結果
4.3.3 結論
4.4 保留遺失音框編碼之音高增益資訊之訊號重整
4.4.1 研究動機
4.4.2 實驗結果
4.4.3 結論
4.5 保留遺失音框LSF係數和音高增益資訊之訊號重整
4.5.1 研究動機
4.5.2 實驗結果
4.5.3 結論
4.6 聲音壓縮管理員
4.6.1 聲音壓縮管理員簡介
4.6.2 工作原理
4.6.3 ACM語音編碼器實作
4.6.4 安裝使用方法
4.6.5 ACM語音編碼器的及時語音處理
4.6.6 結論
4.7 DSP IC硬體的模擬
第五章 結論與未來展望
參考文獻
授權書

參考文獻
[1]Bishnu S. Atal, V. Cuperman and A. Gersh, Advances in Speech Coding, Kluwer-Academic Publishers, 1991.
[2] Bishnu S. Atal, “Predictive Coding of Speech at Low Bit Rates”, IEEE Transaction on Communications Vol. COM-30 NO. 4, Apr. 1982.
[3] Bishnu S. Atal, “High Quality Speech at Low Bit Rates: Multi-Pulse and Stochastically Excited Linear Predictive Coders”, ICASSP, 1986.
[4] J. P. Campbell, V. C. Welch, T. E. Tremain, “An Expandable Error-Protected 4800 bps CELP coder (US Federal Standard 4800 bps voice coder)” IEEE Int. Conf. Acoustics, Speech, and Signal Processing, vol. 2 pp.735-738, 1989.
[5] N. Erdol, C. Castelluccia, and A. Zilouchian, “Recovery of Missing Speech Packets Using the Short-Time Energy and Zero Crossing Measurements” IEEE Trans. Speech and Audio Processing, Vol. 1, No. 3, July 1993.
[6] S. Frassi, A. Dufaux, M. Ansorge, and F. Pellandini, “Efficient Algorithm to Compute LSP Parameters from 10th-Order LPC Coefficient” Proc. IEEE Int. Conf. Acoustics, Speech, and Signal Processing, vol. 3 pp. 1707-1710, 1997.
[7] Gersho and R. M. Gray, Vector Quantization and Signal Compression, Kluwer-Academic Publishers, 1992.
[8]D. J. GoodMan, G. B. Lockhart, and O. J. Wasem, “Waveform Substitution Techniques for Recovering Missing Speech Segments in Packet Voice Communication” IEEE Trans. Acoustics, Speech, and Signal Processing, pp. 1440-1448, Dec. 1986.
[9] R. D. D. Iacovo and D. Sereno, “Embedded CELP Coding for Variable Bit-Rate Between 6.4 and 9.6 kbit/s” Proc. IEEE Int. Conf. Acoustics, Speech, and Signal Processing, pp. 681-684, May 1991.
[10] N. S. Jayant and S. W. Christensen, “Effect of Packet Losses in Waveform Coded
Speech and Improvements Due to an Odd-Even Sample-Interpolation Procedure” IEEE Trans. Commun., Vol. COM-29, pp. 101-109, Feb. 1981.
[11] N. S. Jayant and Peter Noll, Digital Coding of Waveforms, Prentice-Hall, 1984.
[12] Suzuki J. and Taka M., “Missing Packet Recovery Techniques for Low-Bit-Rate
Coded Speech” IEEE J. Sel. Areas Commun.,1989, SAC-7, pp. 707-717.
[13]A. M. Kondoz, Digital Speech, John Wiley & Sons, Inc., 1994.
[14] Peter Kabal and Ravi Prakash Ramachandran, “The computation of Line Spectral
Frequencies Using Chebyshev Polynomials” IEEE Transaction on ASSP, Vol. ASSP-34, No.6, Dec. 1986, 1419-1426.
[15] David P. Kemp, Retha A. Sueda, and Thomas E. Tremain, “An Evaluation of 4800
bps Voice Coders”, U.S. Government.
[16] A. V. McCree and T. P. Barnwell III, “A Mixed Excitation LPC Vocoder Model for
Low Bit Rate Speech Coding” IEEE Trans. Speech and Audio Processing, Vol. 3, pp. 242-250, July 1995.
[17] A. V. McCree, K. Truong, E. B. George and T. P. Barnwell, “A 2.4 kbit/s MELP
Coder Candidate for the New U. S. Federal Standard” Proceedings of ICASSP, IEEE, pp. 200-203, May 1996.
[18] J. S. Marques, I. M. Trancoso, J. M. Tribolet and L. B. Almedia, “Improved Pitch
Prediction with Fractional Delays in CELP Coding” IEEE Int. Conf. Acoustics, Speech, and Signal Processing, vol. 2 pp.665-668, 1990.
[19] A. V. Oppenhein and R. W. Schafer, Discrete-Time Signal Processing, Prentice-
Hall, Inc., 1989.
[20] K. K. Paliwal and B. S. Atal, “Efficient Vector Quantization of LPC Parameters at
24 Bits/Frame” IEEE Trans. Speech and Audio Processing, Vol. 1, pp. 3-14, Jan. 1993.
[21] L. R. Rabiner and R. W. Schafer, Digital Processing of Speech Signals, Prentice-
Hall Inc., 1978.
[22] M. R. Schroeder, and B. S. Atal, “Code-Excited Linear Prediction (CELP):High-
Quality Speech at Very Low Bit Rates” Proc. IEEE Int. Conf. Acoustics, Speech, and Signal Processing, pp. 937-940, Mar. 1985.
[23] I. M. Trancoso and B. S. Atal, “Efficient Procedures for Finding the Optimum
Innovation in Stochastic Coders”, ICASSP Apr. 8-11 1986.
[24] N. Y. Yin, “Congestion Control for Packet Voice by Selective Packet Discarding”
IEEE Trans. Comm., pp. 674-683, May 1990.
[25] Mei Yong, “Study of Voice Packet Reconstruction Methods Applied to CELP
Speech Coding” Proc. IEEE Int. Conf. Acoustics, Speech, and Signal Processing, Vol-2, pp. 125-128, Mar. 1992.
[26] 謝依蘭 , 語音訊號數位處理, 松崗電腦圖書資料股份有限公司,1993.
[27] Frank K. Soong and Biing-Hwang Juang, “Line Spectrum Pair(LSP) and Speech
Data Compression” ICASSP 1.10.1 — 1.10.4 1984.
[28] Yu-Hung Kao, “Low Complexity CELP Speech Cdoing at 4.8 kbps” Thesis of the
Graduate School of the University of Maryland 1990.