Piotr Żelasko
Piotr Żelasko
I'm going to use @danpovey's solution rather than @csukuangfj's solution -- unfortunately, it is not straightforward to estimate how many utterances should be dropped in `partition_cut_ids`, since we have a...
@danpovey @csukuangfj can you please try out the version in PR https://github.com/lhotse-speech/lhotse/pull/267 and let me know if it helped? I won't be able to test the snowfall distributed training setup...
Merged!
FYI this could be of interest to us https://huggingface.co/blog/accelerate-library
> @csukuangfj I'll talk to Kangwei about doing this, it would be a good first project that can let him understand the basics of k2 C++ programming. Cool! In that...
Could be a good opportunity to get more familiar with k2's C++ code. I'll start with the Python part and let's see then.
@jtrmal I won't find the time to work on it this week -- if you want to, feel free to start (just let me know if you do).
Nice! I think there are two more easy wins: using tglarge for decoding (I think we’re using tgmed currently) and saving checkpoints more frequently than per epoch so we can...
The data augmentation setup probably needs some tuning. I ran the full libri recipe as-is, and got: ``` Epoch 3: 2021-04-07 11:42:46,500 INFO [common.py:357] [test-clean] %WER 5.53% [2909 / 52576,...
I think we already have implemented "both" sides padding; which would center the cuts. It'd look sth like this (except the cuts would be concatenated first, with data augmentation applied):...