The authors conduct large-scale experiments on mixed-bandwidth DNN acoustic modeling for ASR using 1,150 hours of wideband and 2,300 hours of narrowband data, testing downsampling, upsampling, and bandwidth extension strategies on 8 test sets with distributed GPU training.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
eess.AS 1years
2019 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Large-Scale Mixed-Bandwidth Deep Neural Network Acoustic Modeling for Automatic Speech Recognition
The authors conduct large-scale experiments on mixed-bandwidth DNN acoustic modeling for ASR using 1,150 hours of wideband and 2,300 hours of narrowband data, testing downsampling, upsampling, and bandwidth extension strategies on 8 test sets with distributed GPU training.