课程名称︰数位语音处理概论 课程性质︰选修 课程教师︰李琳山 开课学院：电资学院 开课系所︰资工系 考试日期（年月日）︰2005.12.2 考试时限（分钟）：120 是否需发放奖励金：是 （如未明确表示，则不予发放） 试题 : #OPEN EVERYTHING #除专有名词可用英文以外，所有文字说明一律以中文为限，未用中文者不计分 #Total points: 120, Time allocation: 1point/min ----------------------------------------------------------------------- 1.(10)Explain the concept of "Corpus-based Text-to-Speech Synthesis", how it works and why it is good. _ 2.(20)Given a HMM λ=(A,B,π), an observation sequence O = o1o2...ot...oT _ and a state sequence q = q1q2...qt..qT, define αt(i) = Prob[o1 o2...ot, qt = i|λ] βt(i) = Prob[ot+1 ot+2 ... oT|qt = i, λ] αt(i)βt(i) (a)(10)Let γt(i) = --------------------- , Explain what γt(i) is, where N is Σi=1toN αt(i)βt(i) total number of states. (b)(10)Formulate and describe the Viterbi algorithm to find the best state _* * * * * _* _ sequence q = q1q2...qt...qT giving the highest probability Prob(q , O|λ). Explain how it works. 3.(10)What is LBG algorithm and why is it better than K-means algorithm? 4.(10)Explain why and how the unseen triphones can be trained using decision trees. 5.(10)Explain the meaning of the perplexity of a language model with respect to a testing corpus. 6.(10)Explain the principles and procedures of estimating the probabilities for unseen events in Katz smoothing. 7.(10) (a)(5)What are the voiced/unvoiced speech signals and their time-domain wave form characteristics? (b)(5)What is pitch in speech signals and how is it related to the tones in Mandarin Chinese? 8.(10)Write down what you know about the techniques for speech end point detection. 9.(10)Explain how the tree lexicon can be used in the search algorithm for large vocabulary continuous speech recognition and how it is helpful. 10.(20)Write down anything you learned about the following subjects which were NOT mentioned in the class. Don't write anything mentioned in the class. (a)Conversational interfaces (b)Search problem/algorithms for large vocabulary continuous speech recognition (sol) 2-a αt(i)是给定modelλ，看到observation o1o2...ot且在时间t时的state为i的机率。 βt(i)是给定modelλ，时间t时在state i的机率，而 ^ αt(i)βt(i) P(O, qt = i|λ) γt(i) = ---------------------- = ---^-------------- Σj=1toN αt(j)βt(j) P(O|λ) 故γt(i)是给定一observation sequence o1o2...oN时间t时state为i的机率。 2-b 定义一新变数δt(i)其值为时间t时到达state i的单一路径之中最高的机率。则δt(i) 数学定义为: δt(i) = max P[q1,q2,...,qt-1,qt = i, o1,o2,...,ot|λ] q1*q2*...qN* 则我们可以进一步得到一递回定义: δt+1(j) = max[δt(i)aij]‧bj(ot+1) i 故Viterbi演算法利用此递回定义配合Dynamic Programming可快速找出最佳值并由 Backtracking找到最佳路径。 4 Decision tree可以根据一连串的feature及利用entropy为splittingcriteria来有效分 类资料。 在Triphone的资料里，许多的event在training data之中会成为unseenevent因此造成 training上的困难。因此根据声学的知识，我们可以将各式各样的triphone分类，使得 在发声上接近的triphone能够被找出，让unseentriphone能够由类似的triphone得到 适当的资料来估计。 6 Katz Smoothing是由Good-Turing Smoothing而来，後者会将高frequency的event之机率 搬动挪作低frequency的event的机率，但此一作法是设法总是有更高frequency的event 存在。在实际应用时此一假设若不成立则会遗失最高frequency event的机率。同时高 frequency的event往往是可信任的，因此Katz Smoothing真对frequency低於某一 threshold的eent做smoothing，若高於此一threshold则不调整。将低於threshold的 event做discount，将frequency分给unseen events,再依照Good-Turing Smoothing。 7-a voiced signals:由振动声代所产生的声音，称为浊音。在time domain上的波形特徵是会 有固定的形状重复出现。 unvoiced signals:不经由振动声带所产生的声音，称为清音。在time domain上的波形 特徵是没有固定的形状重复出现。 7-b pitch指的是声音的音高，频率越高的声调越高。在中文里经由变化音高造成声调，倒是 一声是音高不变，二声则是音高从低变高，三声则是由高变低再拉高。 9 tree lexicon的做法是将每个字的各声学单元视为一个node，则在辨识时对每node只需 找出其最佳path即可得知到达该node的最大机率。这样的好处是可以大量节省空间的使用 ，并且可以加快运算速度，但是因为会舍弃掉一部分资讯，因此不会找到最佳解。 --

※ 发信站: 批踢踢实业坊(ptt.cc) ◆ From: 218.167.77.218