githubharald/CTCDecoder
Connectionist Temporal Classification (CTC) decoding algorithms: best path, beam search, lexicon search, prefix search, and token passing. Implemented in Python.
PythonMIT
Issues
- 2
beam search
#27 opened by fangshuiyun - 2
beam_search with a beam_width=1
#26 opened by skalinin - 5
How get probability
#25 opened by kolesnevg - 1
Is support BPE token?
#24 opened by dangvansam - 2
About blankIdx
#21 opened by anhtuanluu - 7
question about prefix beam search
#22 opened by HaoDreamlong - 2
- 1
- 1
Test custom image and word
#18 opened by abdou31 - 1
- 2
- 2
best_path is better than beam_search in my experiments,is there something wrong with my network?
#14 opened by shaoming20798 - 4
Character level probability
#13 opened by meet-minimalist - 4
- 2
Question about language model initialization
#11 opened by xuannianz - 3
- 3
Support for K-Gram LM where k > 2?
#9 opened by Sushil-Thapa - 1
beam_search.py don't support batch data
#8 opened by huangnengCSU - 1
No module named 'editdistance'
#6 opened by anguslou - 4
Language model at word level
#3 opened by marcoleewow - 2
tensor flow op
#4 opened by eliaho - 1
CTC Token Passing
#2 opened by wellescastro - 7