Comments (2)
Thank you very much for your advices.
from kenlm-training.
Hi @IchigoKurosakii
Yes, the transliteration is common for languages such as Arabic. Incidentally, I was also working with Arabic language for quite some time and used a sophisticated transliteration and normalization technique to make the model training possible. So, I understand what your problem is about.
As for the lower casing issue, I don't think you need to lowercase it during pre-processing. I just happen to use lowercasing because that is simpler.
So, you can use your fancy pre-processing (transliteration, normalization etc) in preprocess.py
:
for line in sys.stdin:
for sentence in sentences:
# your fancy preprocessing steps goes here
# e.g. transliteration, normalization
# print(preprocessed_sentence)
Having said that please make sure that your corpus is large enough. Else you might face issues with the model being not powerful enough. [zero probabilities are handled here with the smoothing technique called Kneser-Ney].
Also, since the trained LM was intended to be used with DeepSpeech during decoding, it is expected that DeepSpeech output also spits out the same encoding (transliterated & normalized version of Arabic) as the Language Model data was trained with. Else it won't score it properly (and essentially doesn't make sense to use this LM during decoding anyway!)
from kenlm-training.
Related Issues (7)
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from kenlm-training.