課程名稱︰數位語音處理概論
課程性質︰選修
課程教師︰李琳山
開課學院：電資學院
開課系所︰電機、資工系
考試日期（年月日）︰2007.06.26
考試時限（分鐘）：120
是否需發放獎勵金：是
（如未明確表示，則不予發放）
試題 :
Digital Speech Processing, Final Exam
June. 26, 2007, 10:10-12:10
● OPEN EVERYTHING
● 除專有名詞可用英文以外，所有文字說明一律以中文為限，未用中文者不計分
● Total points: 160
● Note that you don't need to be able to answer all the questions.
───────────────────────────────────────
1. (20) In eigenvoice approach, explain how the eigenvoice space is
constructed, what that means, and why rapid speaker adaptation can be
achieved with very limited quantity of data?
2. (20) In Latent Semantic Analysis the elements w_ij of the word-document
_
matrix W is
c_ij
w_ij = (1 - ε_i) * ─── ,
n_j
Where c_ij is the number of times the word w_i occurs in the document d_j,
n_j is the total number of words in d_j, and
1 N ┌ c_ij ┐ ┌ c_ij ┐ N
ε_i = - ─── * Σ│───│log│───│ , t_i = Σ c_ij ,
log(N) j=1└ t_i ┘ └ t_i ┘ j=1
where N is the total number of documents. Explain the meaning of all these
parameters.
3. (20) For Linear Discriminative Analysis (LDA), explain the meaning of
within/between class scatter matrices S_W/S_B, and the meaning of the
optimization criterion.
4. (20) Explain why and how vector space model based on subword units are
useful in retrieving speech information using speech queries.
5. (20) Write down the two steps in each iteration of EM algorithm and explain
how they operate and what they mean.
6. (20) Write down what you know about likelihood ratio test and its
applications in digital speech processing.
7. (20) Explain the three key elements in a spoken dialogue system: what they
are, how they operate, and how they are linked together.
8. (20) Compare the client-server model and server-only model for Distributed
Speech Recognition (DSR), and discuss the different considerations.