A mini conference on synthetic voices.
Chenzi Xu · chenzi.xu@ntu.edu.sg
(a)(i) Fill in the Viterbi table. The word ends after frame 3, so do not count the transition out of /i/. [2]
(a)(ii) Give the best path and its probability. [1]
(b) When several routes arrive at the same cell, what do Viterbi and the CTC forward table put there, and why? [2]
(c)(i) Forced alignment fills the same kind of table, with the transcript already known. What does it already know, and what does it give you? [1]
(d) All three methods fill a table. What is the benefit of that? [1]
V(s, t): the probability of the best path that ends in state s at frame t
a route: one cell one frame back, times the transition out of it
max: keep the biggest route, remember where it came from, discard the rest
● /s/ ● /i/ ● the route kept, and the best path ● the route discarded
Comparing the cells instead of the routes gives 0.0216 and the path /s/ → /i/ → /i/. It was the most common wrong answer, and it scored nothing on (a)(ii).
V(/s/, 3) scores the best route into that cell, and a cell knows nothing about what follows it. The end-in-/i/ rule applies when you read the answer off, not while you fill the table: with a fourth frame, /s/ is live again.
An arc from /i/ back to /s/ would mean the pronunciation went /s i s/, which is a different word. The rows also leave no room for it: 0.4 + 0.6 and 0.3 + 0.7 already come to 1.
Hot and cold can follow each other in any order, so every arc exists and the matrix is full. Nothing fixes the sequence in advance.
The same recursion, the same table, the same backtrace. Only the set of allowed arcs differs, and that is what the word's spelling buys you.
Nothing is approximated: what comes next depends only on the state a path is in, not on the way in, so a dropped route could never have won later.
Adding costs and keeping the smallest picks the same route as multiplying probabilities and keeping the largest. Unit selection keeps the best route like Viterbi; it just counts in costs, and any recording can follow any other.
Panel sharing + Q&A
Question starters, if you need one:
21 posters · 3 minutes each · quick rotation
To Ezio, to all 21 presenters, and to all of you.
Tonight, by 23:59: abstract and poster on NTULearn