CASMACAT: A Computer-assisted Translation Workbench

Abstract

CASMACAT is a modular, web-based translation workbench that offers advanced functionalities for computer-aided translation and the scientific study of human translation: automatic interaction with machine translation (MT) engines and translation memories (TM) to obtain raw translations or close TM matches for conventional post-editing; interactive translation prediction based on an MT engine's search graph, detailed recording and replay of edit actions and translator's gaze (the latter via eye-tracking), and the support of e-pen as an alternative input device.The system is open source sofware and interfaces with multiple MT systems.

Development of English-Hindi Interactive Machine Translation

“…Similar approaches we used by Goyal et al [58] [59] for their models. CASMACAT (Alabau et al, 2014) [60], a modular, web-based translation workbench with advanced functionality and editing features provided us with TM for raw match and automatic translations from MT server for post editing, consisting of a GUI, backend, an MT server and a CAT server.…”

“…of words in the documents. This is shown in equation (1). Here the calculated speed of doing translations when no suggestions were provided and when the suggestions were provided.…”

“…Section 4 evaluates the system and Section 5 concludes the same. Nagao (1984) [1] proposed the idea of machine translation based on the examples, which were made available to the system prior to the translation process. A bilingual corpus was used which was comprised of source language text and their corresponding target language text.…”

A Simple and Effective Approach to Automatic Post-Editing with Transfer Learning

“…APE is particularly appealing for rapidly customizing MT, avoiding to train new systems from scratch. Interfaces where human translators can post-edit and improve the quality of MT sentences (Alabau et al, 2014;Federico et al, 2014;Denkowski, 2015;Hokamp, 2018) are a common data source for APE models, since they provide triplets of source sentences (src), machine translation outputs (mt), and human post-edits (pe).…”

Statistical Modelling and Automatic Tagging of Human Translation Processes

“…The increasing popularity of PE in the translation industry has enticed researchers in MT to try to find new ways of using MT to make human translation faster and less cognitively demanding. Alabau et al (2014), for example, use interactive MT to provide translators with automatic sentence completion, and they use automatic MT quality estimation to identify and highlight parts in the MT output that are likely to be of poor quality and therefore probably need revision (see also Chaps. 3,4,5,and 10).…”