Open source challenger takes on Google Translate--independant Indian language translation initiatives can adapt OpenNMT
3 views
Skip to first unread message
Ashok.S. Subramanian
Jan 1, 2017, 10:50:43 PM1/1/17
to Senthil Nathan (Aazhi), Anand G., Deepak Pawar, Garga Chatterjee, Ganesh Chetan, Joga Virk, Mani Manivannan, Priyank KS, Sakethsahu, Vasanth Shetthy, Vivek V., Malayala Aikya Vedi, Pavithran Dr., Prof.V.P.Markose Vpmarkose, R.Nandakumar Nanda, Subair K.K., Sureshputhenparambil P., Saju Kochery, Bharathan Km, Dr.Bichu.X.Malayil, Shiju R., DevesanPerur, Benny.P.Neeleswaram, Priyesh Palakkad, Harikumar Palakkad, BBP Bharatheeya Bhasha Prasthanam, Shinoj Kozhikode, SunilPElayidom, ramank...@gmail.com, Soumya Thomas
Open source challenger takes on Google Translate
There's a new open source machine translation framework, but Google with its trove of language data could have the upper hand
InfoWorld | Dec 20, 2016Credit: CSO staff By Serdar Yegulalp | Senior Writer
Researchers have released an open source neural network system for performing language translations that could be an alternative to proprietary, black-box translation services. Open Source Neural Machine Translation (OpenNMT) merges work from researchers at Harvard with contributions from long-time machine-translation software creator Systran. It runs on the Torch scientific computing framework, which is also used by Facebook for its machine learning projects.
Ideally, OpenNMT could serve as an open alternative to closed-source projects like Google Translate, which recently received a major neural-network makeover to improve the quality of its translation.
But the algorithms aren't the hard part; it’s coming up with good sources of data to support the translation process—which is where Google and the other cloud giants that provide machine translation as a service have the edge.
Speaking in tongues
OpenNMT, which uses the Lua language to interface with Torch, works like other products in its class. The user prepares a body of data that represents the two language pairs to be translated—typically the same text in both languages as translated by a human translator. After training OpenNMT on this data, the user can then deploy the resulting model and use it to translate texts.
There's a new open source machine translation framework, but Google with its trove of language data could have the upper hand
InfoWorld | Dec 20, 2016Credit: CSO staff By Serdar Yegulalp | Senior Writer
Researchers have released an open source neural network system for performing language translations that could be an alternative to proprietary, black-box translation services. Open Source Neural Machine Translation (OpenNMT) merges work from researchers at Harvard with contributions from long-time machine-translation software creator Systran. It runs on the Torch scientific computing framework, which is also used by Facebook for its machine learning projects.
Ideally, OpenNMT could serve as an open alternative to closed-source projects like Google Translate, which recently received a major neural-network makeover to improve the quality of its translation.
But the algorithms aren't the hard part; it’s coming up with good sources of data to support the translation process—which is where Google and the other cloud giants that provide machine translation as a service have the edge.
Speaking in tongues
OpenNMT, which uses the Lua language to interface with Torch, works like other products in its class. The user prepares a body of data that represents the two language pairs to be translated—typically the same text in both languages as translated by a human translator. After training OpenNMT on this data, the user can then deploy the resulting model and use it to translate texts.
.....................................
....................................................
............................
.............................................
|
Reply all
Reply to author
Forward
0 new messages