iamanigeeit
iamanigeeit
@wanshun123 Did you train using `use_gst=False`? I have the same issue when `use_gst=False` but not when `True`. @syang1993 In my case the audio seems intelligible, although not good quality. I...
@issakh Global unstructured pruning is supported by Pytorch. See [the tutorial](https://pytorch.org/tutorials/intermediate/pruning_tutorial.html). The only reason we do structured pruning is to reduce model size overall. Unstructured pruning is only masking.
@issakh Heh, lots of people ask this as well, [see here](https://discuss.pytorch.org/t/global-structured-pruning/67263) The question is how do you adjust the number of parameters to prune for each layer when the layer...
@kan-bayashi Sorry for the delay because of ICASSP submission. I will pull the latest version of ESPnet and fix my PR this week!
@obones I'm using this conversion for my own work. (`UW` without any number seems to be a mistake -- CMU dictionary does not have any entry with `UW`) ``` {...
Hello @kan-bayashi I have made the requested changes, but i can't submit the PR, because i have modified other files for my own work on `iamanigeeit:master`. I tried to fork...
Thanks @kan-bayashi Creating a new branch and making the PR from Github instead of Pycharm seems to work. The new commit is at https://github.com/espnet/espnet/pull/4801
@hertz-pj This is old, but just putting it there in case someone is searching for a comparison. If you want to compare inference only, you can simple download pretrained models...
@auspicious3000 This works well enough for me. ``` import regex as re from pathlib import Path from phonemizer.backend import BACKENDS from phonemizer.separator import Separator def espeak_english_replacements(word_sep): rules_file = Path("espeak-ng") /...
@migi-gon It seems your model is not learning at all. After 50 epochs there should be intelligible speech, although the multi-speaker won't work well (if at all). But i think...