擾動輔助EMD演算法在MCU中即時語音處理之評估

以作者查詢圖書館館藏

、以作者查詢臺灣博碩士

、以作者查詢全國書目

、勘誤回報

、線上人數：13

、訪客IP：13.59.212.110

姓名

陳楷仁(Kai-Ren Chen) 查詢紙本館藏

畢業系所

機械工程學系

論文名稱

擾動輔助EMD演算法在MCU中即時語音處理之評估
(The Evaluation of Real-time Speech Signal Processing with the Disturbance-Assisted EMD Algorithm on an MCU)

相關論文

★ 二十一點遊戲之正確期望值模型：以遞迴之方式實行隨機訊號處理	★ 擾動輔助經驗模態分解之邊界效應的理論與數值分析
★ 基於加速度計的高精度步數演算法	★ 基於加速度計低功耗精確量測步數演算法在穿戴式裝置的實現
★ 經驗模態分解局部性之近場性質的一些證明	★ 擾動輔助EMD演算法在穿戴式嵌入式裝置中MCU的即時運算
★ 總體相位經驗模態分解	★ 利用多孔介質提升甲烷熱裂解產氫之研究

檔案

[Endnote RIS 格式]

[Bibtex 格式]

[相關文章]

[文章引用]

[完整記錄]

[館藏目錄]

[檢視]

[下載]

本電子論文使用權限為同意立即開放。
已達開放權限電子全文僅授權使用者為學術研究之目的，進行個人非營利性質之檢索、閱讀、列印。
請遵守中華民國著作權法之相關規定，切勿任意重製、散佈、改作、轉貼、播送，以免觸法。

摘要(中)

經驗模態分解能根據訊號本身的特性提取具有物理意義的本質模態函數。因此在處理複雜的語音訊號時，經驗模態分解具有顯著的優勢。然而經驗模態分解提取的本質模態函數經常出現模態混合現象，導致其物理意義受損。因此許多擾動輔助經驗模態分解被提出，例如總體經驗模態分解與均勻相位經驗模態分解，來改善此問題。儘管擾動輔助經驗模態分解在處理非線性及非平穩訊號方面具有優勢，但其高運算量與記憶體需求對資源有限的嵌入式系統構成挑戰。在本研究將評估擾動輔助經驗模態分解演算法在嵌入式穿戴裝置中常用的微控制器中即時處理語音訊號的可行性，並在符合即時語音處理限制的條件下進行最佳化以降低計算時間、記憶體需求與系統延遲。經過最佳化，研究結果顯示，使用即時自適應擾動輔助經驗模態分解演算法進行語音除噪時，對於採樣頻率為16k Hz的語音訊號，計算負載比達54%、音訊延遲為0.03秒，記憶體需求為18.75 KB；而進行語音特徵處理時，對於8k Hz的語音訊號，計算負載比為59.67%、音訊延遲為0.1697秒，記憶體需求為68.75 KB。雖然在微控制器上實現以擾動輔助經驗模態分解進行即時語音處理具有可行性，但計算量與音訊延遲依舊是很大的問題。

摘要(英)

Empirical Mode Decomposition (EMD) extracts intrinsic mode functions (IMFs) with physical significance based on the signal′s characteristics, making it advantageous for complex speech signal processing. However, EMD often suffers from mode mixing, which undermines its physical interpretation. To address this, disturbance-assisted EMD (DA-EMD) methods like Ensemble EMD (EEMD) and Uniform Phase EMD (UPEMD) have been proposed. While DA-EMD excels in handling nonlinear, non-stationary signals, its high computational load and memory requirements pose challenges for resource-limited embedded systems. This study evaluates the feasibility of implementing DA-EMD for real-time speech processing on microcontrollers (MCUs) in embedded wearable devices, optimizing it to reduce computation time, memory usage, and system latency. The optimized Real-Time UPEMDA algorithm demonstrated a 54% computational load, 0.03-second audio delay, and 18.75 KB memory usage for 16kHz speech denoising. For 8kHz speech feature extraction, the computational load was 59.67%, with a 0.1697-second delay and 68.75 KB memory usage.

關鍵字(中)

★ 經驗模態分解
★ 語音處理
★ 即時運算
★ 微控制器

關鍵字(英)

★ EMD
★ Speech Processing
★ Real-Time Computing
★ MCU

論文目次

中文摘要 i
英文摘要 ii
誌謝 iii
目錄 iv
圖目錄 vii
表目錄 viii
符號說明 ix
一、緒論 1
1-1研究動機 1
1-2研究目的 1
二、經驗模態分解的基礎 3
2-1三次木條曲線內插 3
2-1-1邊界條件 4
2-1-2三對角矩陣線性方程的解 6
2-2經驗模態分解 7
2-2-1經驗模態分解演算法 7
2-2-2計算複雜度與記憶體複雜度 11
2-2-3低記憶體經驗模態分解 13
2-3擾動輔助經驗模態分解 14
2-3-1總體經驗模態分解與互補總體經驗模態分解 14
2-3-2均勻相位經驗模態分解 16
2-3-3自適應均勻相位經驗模態分解 18
三、即時語音處理 11
3-1語音處理 19
3-2即時運算 20
3-3最佳化即時語音處理之成本 21
3-3-1最佳化計算時間 21
3-3-2最佳化記憶體使用量 21
3-3-3最佳化音訊延遲 21
3-4即時語音處理限制 21
3-4-1即時處理限制 21
3-4-2記憶體限制 21
3-4-3經驗模態分解視窗限制 21
3-4-4訊號連續限制 21
3-4-5音訊延遲限制 21
3-5語音除噪測試 23
四、嵌入式穿戴裝置評估 29
4-1微控制器規格 29
4-2單位篩選時間與計算時間 29
4-3擾動輔助經驗模態分解演算法處理語音訊號的成本 31
4-3-1語音除噪 31
4-3-2語音特徵擷取 33
五、結論與未來展望 35
5-1結論 35
5-2未來展望 36
參考文獻 37

參考文獻

〔1〕 Lee, W. T., “Tridiagonal matrices: Thomas algorithm,” MS6021, Scientific Computation, University of Limerick, 2011.
〔2〕 Huang, Norden E., et al. “The empirical mode decomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysis,” Proceedings of the Royal Society of London. Series A: Mathematical, Physical and Engineering Sciences, vol. 454, no. 1971, pp. 903-995, 1998.
〔3〕 Wang, Yung-Hung, et al., “On the computational complexity of the empirical mode decomposition algorithm. ” Physica A: Statistical Mechanics and its Applications, vol. 400, pp. 159-167, 2014.
〔4〕 Wang, Y. H., Lin, Y. C., “On the memory cost of EMD algorithm.” IEEE Access, vol. 10, pp. 114242-114251, 2022.
〔5〕 Huang, Norden E., Zheng Shen, and Steven R. Long, “A new view of nonlinear water waves: the Hilbert spectrum,” Annual Review of Fluid Mechanics, vol. 31, no. 1, pp. 417-457, 1999.
〔6〕 Wu, Zhaohua, and Norden E. Huang, “Ensemble empirical mode decomposition: a noise-assisted data analysis method,” Advances in Adaptive Data Analysis, vol. 1, no. 1, pp. 1-41, 2009.
〔7〕 Yeh, J.-R., J.-S. Shieh, and N. E. Huang, “Complementary ensemble empirical mode decomposition: A novel noise enhanced data analysis method,” Advances in Adaptive Data Analysis, vol. 2, no. 2, pp. 135-156, 2010.
〔8〕 Wang, Yung-Hung, Kun Hu, and Men-Tzung Lo, “Uniform phase empirical mode decomposition: An optimal hybridization of masking signal and ensemble approaches,” IEEE Access, vol. 6, pp. 34819-34833, 2018.
〔9〕 Colominas, Marcelo A., Gaston Schlotthauer, and Maria E. Torres, “Improved complete ensemble EMD: A suitable tool for biomedical signal processing,” Biomedical Signal Processing and Control, vol. 14, pp. 19-29, 2014.
〔10〕 Lin, L., and J. Hongbing, “Signal feature extraction based on an improved EMD method,” Measurement, vol. 42, no. 5, pp. 796-803, 2009.
〔11〕 Kerkeni, L., Serrestou, Y., Raoof, K., Mbarki, M., Mahjoub, M. A., and Cleder, C., “Automatic speech emotion recognition using an optimal combination of features based on EMD-TKEO,” Speech Communication, vol. 114, pp. 22-35, 2019.
〔12〕 Khaldi, K., Boudraa, A. O., Bouchikhi, A., and Alouane, M. T. H., “Speech enhancement via EMD,” EURASIP Journal on Advances in Signal Processing, pp. 1-8, 2008.
〔13〕 Chatlani, N., and Soraghan, J. J., “EMD-based filtering (EMDF) of low-frequency noise for speech enhancement,” IEEE Transactions on Audio, Speech, and Language Processing, vol. 20, no. 4, pp. 1158-1166, 2011.
〔14〕 Zao, L., Coelho, R., and Flandrin, P., “Speech enhancement with EMD and Hurst-based mode selection,” IEEE/ACM Transactions on Audio, Speech, and Language Processing, vol. 22, no. 5, pp. 899-911, 2014.
〔15〕 Chang, N., T. Chen, C. Chiang, and L. Chen, “On-line empirical mode decomposition biomedical microprocessor for Hilbert Huang transform,” in Proceedings of the IEEE Conference on Biomedical Circuits and Systems, pp. 420-423, 2011.
〔16〕 Faltermeier, R., et al., “Weighted sliding empirical mode decomposition,” Advances in Adaptive Data Analysis, vol. 3, no. 4, pp. 509-526, 2011.
〔17〕 Faltermeier, R., A. Zeiler, I. R. Keck, A. M. Tome, A. Brawanski, and E. W. Lang, “Sliding empirical mode decomposition,” in Proceedings of the International Joint Conference on Neural Networks, pp. 1-8, 2010.
〔18〕 Fontugne, R., P. Borgnat, and P. Flandrin, “Online empirical mode decomposition,” in Proceedings of the IEEE International Conference on Acoustics, Speech and Signal Processing, pp. 4306-4310, 2017.
〔19〕 CY, C., & CMY, K. ITU-T Standards and Recommendations G.114.
〔20〕 Wang, Y. H., Liang, S. F., Kuo, T. B., and Lin, Y. C., “Software Implementation of Real-time EMD-based Algorithm in Embedded Microprocessors for Wearable Devices,” IEEE Transactions on Instrumentation and Measurement, 2024.
〔21〕 Wang, Y.-H., I.-Y. Chen, H. Chiueh, and S.-F. Liang, “A low-cost implementation of sample entropy in wearable embedded systems: An example of online analysis for sleep EEG,” IEEE Transactions on Instrumentation and Measurement, vol. 70, pp. 1-12,. 2021.
〔22〕 Hsieh, T.-H., M.-H. Liu, C.-E. Kuo, Y.-H. Wang, and S.-F. Liang, “Home use and real-time sleep-staging system based on eye masks and mobile devices with a deep learning model,” Journal of Medical and Biological Engineering, vol. 41, no. 5, pp. 659-668, 2021.

指導教授

王淵弘(Yung-Hung Wang)

審核日期

2024-12-30

推文