針對智慧型輪椅之語音辨識與控制之實現

The Implementations of Voice Recognition and Control for Intelligent Wheelchair

廖亦珉(Yi-Min Liao)

指導教授 ： 蔡舜宏

共同指導教授 ： 陳亮嘉

國立臺北科技大學/機電學院/自動化科技研究所/碩士(2014年)

https://doi.org/10.6841/NTUT.2014.00787

摘要

本論文基於Raspberry Pi嵌入式平台開發一套智慧型輪椅語音辨識與控制系統，搭配外部立體聲麥克風以及USB音效卡構成系統之硬體架構應用於電動輪椅上。透過使用者的聲音經由麥克風收取聲音訊號，並將語音訊號進行語音前處理再求取訊號之特徵參數。此外，透過模糊C平均值演算法(Fuzzy c-means)進行語音分群比對，使其能準確辨識出特定使用者之語意以決定輸出電壓之大小以控制電動輪椅。最後，經由實驗結果可驗證所提出之智慧型輪椅語音辨識系統之精確性及可靠性。

關鍵字

智慧型輪椅 ； 語音分群 ； 嵌入式系統

並列摘要

In this thesis, the voice recognition and control system for intelligent wheelchair based on Raspberry Pi embedded platform is implemented. With the external stereo microphone and USB sound devices in hardware architecture, the uesr’s voice is collected by microphone and transformed into audio signal to execute the preprocess for obtaining the signal characteristic parameter. In addition, Fuzzy c-means clustering algorithm is adopted to classify the voice signal for comparison. The specific users’ semanteme can be recognized accurately and the voltage can be determined for the control of the wheelchair. Finally, an experiment is shown the accurately and reliability of the proposed intelligent wheelchair voice recognition system.

並列關鍵字

Intelligent Wheelchair ； Voice Clustering ； Embedded System

參考文獻

[10] 盧嘉泓(2009)，電動輪椅重要規格整理與比較，元智大學老人福祉科技研究中心。

[13] L. R. Rabiner and M. R. Sambur, “An Algorithm for Determining the Endpoints of Isolated Utterances,” The Bell System Technique Journal, Vol. 54, February 1975, pp. 297-315.

[14] J.R. Deller, Discrete Time Processing of Speech Signals, Macmillan, 1993.

[15] Rabiner, L. and B. H. Juang, Fundamentals of Speech Recognition, Prentice-Hall, 1993.

[16] D. Burshtein, “Robust parametric modeling of duration in hidden Markov models,” IEEE Trans. on Speech Audio Processing, vol. 4, pp. 240-242, May 1996.