Rating 2.9 out of 5 (17 ratings in Udemy)
What you'll learn- Understand the basics of statistical machine translation
- Installing and configuring Moses SMT tool
- Train and test their own phrase based translation model using Moses tool
- Tune translation parameters to find optimal weights for the translation model maximising translation performance
- Learn tuning algorithms
- Train and test their own language model with popular language modelling toolkit IRSTLM and SRILM
- Evaluate their …
Rating 2.9 out of 5 (17 ratings in Udemy)
What you'll learn- Understand the basics of statistical machine translation
- Installing and configuring Moses SMT tool
- Train and test their own phrase based translation model using Moses tool
- Tune translation parameters to find optimal weights for the translation model maximising translation performance
- Learn tuning algorithms
- Train and test their own language model with popular language modelling toolkit IRSTLM and SRILM
- Evaluate their translation system using standard metrics
- How to run machine translation as a service serving various clients
- Optimise moses for faster training, tuning and testing of the translation system
DescriptionMachine Translation is a well-known sub field of Computational Linguistics that investigates the use of software to automatically translate text or speech from one language to another. It becomes a challenge to handle language specific features when it comes to machine translation such as differences in linguistic grammar, translation of idioms, and the isolation of anomalies particular to a language. Statistical and neural methods are being used extensively today to solve this problem and to get better translations. In this course we would focus on building a fast and high performance statistical machine translation system learning the various methodologies, tools and techniques widely used to accomplish it. In this course you will get hands-on in building your own statistical machine translation system using the Open Source Moses framework.
