Full metadata record
DC FieldValueLanguage
dc.contributor.author古鴻炎en_US
dc.contributor.authorGU, HONG-YANen_US
dc.contributor.author李其昌en_US
dc.contributor.authorLI, GI-CHANGen_US
dc.date.accessioned2014-12-12T02:02:57Z-
dc.date.available2014-12-12T02:02:57Z-
dc.date.issued1984en_US
dc.identifier.urihttp://140.113.39.130/cdrfb3/record/nctu/#NT732241011en_US
dc.identifier.urihttp://hdl.handle.net/11536/51959-
dc.description.abstract論文所設計的非特定語者單字語音辨認系統,使用了線性預估編碼(LINEAR PREDCTION CODING),二元檥向量量(BINARY TREE VECTOR QUANTIZATION ),隱 藏式馬可夫模型(HIDDEN MARKOV MODEL )等技術,目的訧是要在合理的辨認準確 度下,盡量縮短辨認時間和所佔的記憶空間,這樣,才能讓系統達到即時(REAL TIME)的要求及做到實用的目標。 除了技術的選擇外,論文也對計算機求值的方法作了考慮,例如:減少運算的次數. 將乘法經由取對數而化成加法等。此外,我們也做了實驗,結果令人相當滿意。zh_TW
dc.language.isozh_TWen_US
dc.subject二元樹向量量化zh_TW
dc.subject線性預估編碼zh_TW
dc.subject隱藏式馬可夫模型zh_TW
dc.subject即時zh_TW
dc.subject計算機zh_TW
dc.subject單字語音辨認系統zh_TW
dc.subjectBTVQen_US
dc.subjectLINEAR-PREDICTION-CODINGen_US
dc.subjectHIDDEN-MARKOV-MODELen_US
dc.subjectREAL-TIMEen_US
dc.title使用二元樹向量量化、隱藏式馬可夫模型的一個中文單字語音辨認系統zh_TW
dc.typeThesisen_US
dc.contributor.department資訊科學與工程研究所zh_TW
Appears in Collections:Thesis