Skip to main content
EURASIP Journal on Audio, Speech, and Music Processing

Table 3 Results for the tonal method with two initialization methods for German

From: Linguistically motivated parameter estimation methods for a superpositional intonation model

 

RMSE

Aprates

Aarates

M-ME

1.48 ± 0.19

0.44 ± 0.05

1.11 ± 0.09

A-ME

1.33 ± 0.14

0.44 ± 0.06

1.69 ± 0.1

T-ME (by M-ME)

1.46 ± 0.27

0.49 ± 0.06

1.27 ± 0.13

T-ME (by A-ME)

1.50 ± 0.21

0.49 ± 0.06

1.35 ± 0.14

T-ME Prototype

 

0.49 ± 0.06

1.53 ± 0.16

  1. The RMSE is given in ST and the rates in commands per second. Standard deviation is included as scattering measure.
Back to article page