用反饋式語音識(shí)別理解方案進(jìn)行漢語短語的識(shí)別理解
CHINESE PHRASE RECOGNITION AND UNDERSTANDING WITH INFORMATION FEEDBACK
-
摘要: 漢語語音理解系統(tǒng)的任務(wù)之一是把語音識(shí)別系統(tǒng)獲得的漢語單音節(jié)轉(zhuǎn)換成正確的漢字、詞,乃至漢語的短語、語句,與語音識(shí)別系統(tǒng)一起,完成一個(gè)語音到文本(speech to text)的轉(zhuǎn)換系統(tǒng)。本文利用一個(gè)閉環(huán)反饋方式漢語語音識(shí)別理解方案,在漢語詞識(shí)別理解的基礎(chǔ)上,進(jìn)一步實(shí)現(xiàn)對(duì)漢語結(jié)構(gòu)性短語的識(shí)別理解,獲得了預(yù)期的結(jié)果。最后本文對(duì)實(shí)驗(yàn)結(jié)果和反饋式語音識(shí)別理解方案進(jìn)行了討論。
-
關(guān)鍵詞:
- 語音識(shí)別; 語音理解; 反饋
Abstract: It is one of the tasks of Chinese speech understanding system that translates Chinese syllable from Chinese speech recognition system into a correct Chinese character, word, and phrase, sentence, and it will implement the task of speech to text altogether with Chinese speech recognition system. This paper makes a system on Chinese phrase understanding with a close ring approach with information feedback, which based on Chinese word recognition system. This system can understand the special constructed phras,a predicted result on this experiment is achieved. Finally, this paper gives a conclusion on the experiment and summary about the approach with information feedback. -
Lee. L. F, et al. A best-first language processing model integrating the unification grammar and[2]Markov language model for speech recognition application. IEEE Trans. on SAP, 1993, 1(2), 221-240.[3]Bahl. L. R, et al. A tree-based statistical language model for natural language speech recognition.[4]IEEE bans. on ASSP, 1989, 37(11), 1001-1008.[5]傅秋良,袁保宗.一種新的語音識(shí)別理解方法.北方交通大學(xué)學(xué)報(bào),1996, 20(2), 131-135.[6][4][7]Fu laiuliang, Yuan Baozong, Lin Biqin. Chinese word recognition and understanding with information feedback, ICSP96, Beijing: Oct. 1996, 737-740. -
計(jì)量
- 文章訪問數(shù): 2227
- HTML全文瀏覽量: 177
- PDF下載量: 408
- 被引次數(shù): 0
下載:
