A symmetric blank-selection method for CTC knowledge distillation lets a student model train without any CTC loss and with no loss in word error rate.
This might be due to the fact that blank distributions induced by CTC training might depend a lot on the data, as blank serves both as a wait and a silence token
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
Analyzing the Importance of Blank for CTC-Based Knowledge Distillation
A symmetric blank-selection method for CTC knowledge distillation lets a student model train without any CTC loss and with no loss in word error rate.