From: A unit selection text-to-speech-and-singing synthesis framework from neutral speech: proof of concept
Note
β
dur(ms)
≤ 2.5
(2.5–4]
> 4
600
9.0
23.5
67.5
1200
0.0
0.2
99.8
300
55.1
29.7
15.2
3.5
8.4
88.1
150
97.8
1.8
0.3
50.1
32.6
17.4