Skip to content
This repository has been archived by the owner on Oct 13, 2022. It is now read-only.

Issues: k2-fsa/snowfall

60 Open 40 Closed
60 Open 40 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Assignee
Filter by who’s assigned
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Issues list

Archiving snowfall repo
#250 opened Oct 13, 2022 by pzelasko
1
How to built ctc_topo with blankid != 0
#249 opened Oct 13, 2022 by Slyne
1
Trace frame-level scores using lattice
#248 opened Sep 10, 2021 by zrsjta
19
Bug in decoder_padding_mask in BPE training
#242 opened Aug 3, 2021 by csukuangfj
Bug in computing encoder padding mask
#240 opened Aug 2, 2021 by csukuangfj
1
Rotary positional embeddings for conformer
#235 opened Jul 21, 2021 by pzelasko
Plan for multi pass n-best rescoring
#232 opened Jul 11, 2021 by danpovey
7
Any suggestions/requests for Lhotse API changes/new features before tutorial deadline?
#226 opened Jul 3, 2021 by pzelasko
CTC training speed question
#220 opened Jun 28, 2021 by yuekaizhang
10
Topology choices
#195 opened May 13, 2021 by danpovey
Inconsistency between the reported loss and the actual loss used for gradient computation
#194 opened May 12, 2021 by pzelasko
2
Building lexicons in Python
#191 opened May 11, 2021 by pzelasko
16
error when PytorchStreamReader failed reading file
#186 opened May 7, 2021 by yujunlhz
2
Have a look at SpeechBrain
#185 opened May 7, 2021 by csukuangfj
1
Why do ctc.py and mmi.py use use_double_scores=True?
#183 opened May 6, 2021 by galv
5
torch.multiprocessing.spawn and changing files
#180 opened Apr 28, 2021 by danpovey
Label smoothing for LF-MMI
#179 opened Apr 28, 2021 by danpovey
14
Convergence & decoding problem
#170 opened Apr 21, 2021 by zhichaowang
16
Conformer & convergence
#169 opened Apr 21, 2021 by danpovey
Speeding up training
#165 opened Apr 16, 2021 by danpovey
@csukuangfj
Initialization of last layer to zero
#161 opened Apr 15, 2021 by danpovey
4
Full-libri default true?
#154 opened Apr 12, 2021 by danpovey
9
DDP address in use
#152 opened Apr 10, 2021 by danpovey
20
Full-librispeech training
#146 opened Apr 3, 2021 by danpovey
27
RuntimeError in ctc_att_transformer_train.py
#142 opened Mar 28, 2021 by csukuangfj
1
5
ProTip! Follow long discussions with comments:>50.