Abstract: The current popular knowledge distillation (KD) methods effectively compress the transformer-based end-to-end speech recognition model. However, existing methods fail to utilize complete ...
"Lossless" and "lossy" are both types of compression. A zipped file is lossless, because when you unzip it, every original bit is still there from the original file. A JPEG, however, is a "lossy" file ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results
Feedback