Ye Bai
Ye Bai
Dear all, I am using knowledge distillation training for ASR with lingvo. However, the distillation_loss (cross entropy between the teacher and the student) increases rather than decrease. I am confused...
And it causes dim mismatch.
Hi authors, Thank you for your fantastic work! I have a small question: In Table 1, FID_k, FID_g of groundtruth are reported. I am a little bit confused with this....
Thank you for your transformer code! When I ran the code, I encountered such issue: Traceback (most recent call last): File "train.py", line 539, in train() File "train.py", line 451,...