. Author manuscript; available in PMC: 2021 Sep 1.

Published in final edited form as: Comput Speech Lang. 2020 Feb 18;63:101077. doi: 10.1016/j.csl.2020.101077

Table 3:

WER results for artificially generated acoustic variability on TEDLIUM

	Time-stretching	Pitch-shifting	VTLN-warping	Global
No adaptation	16.08	37.14	13.52	20.05
Bottom layer	16.41	20.68	13.11	16.27
Top layer	16.88	26.77	13.95	18.37