I recently read Bytenet and Wavenet and I was curious why the first model is not as popular as the second. From my understanding, Bytenet can be seen as a seq2seq model where the encoder and the decoder are similar to Wavenet. Following the trends from NLP where seq2seq models seem to perform better, I find it strange that I couldn't find any paper that compares the two. Are there any drawbacks of Bytenet over Wavenet other than the computation time?
Asked
Active
Viewed 644 times
4
1 Answers
0
My conclusion is the same as yours that there doesn't seem to be any published comparison of the two models. ByteNet is computationally expensive and requires a lot of parameters. WaveNet improves on ByteNet's efficiency, as you mentioned, and I believe that is the main difference.
Brian O'Donnell
- 1,997
- 9
- 23