Samsung Electronics’ Galaxy AI-enabled cell units permit customers to get pleasure from seamless, barrier-free communication in much more international locations. Now supporting Arabic, Indonesian and Russian, Galaxy AI’s Interpreter and Live Translate options have expanded from 13 to 16 accessible languages.
Samsung Research mixed knowledge and cutting-edge expertise. Together with the Mobile eXperience (MX) Business R&D Office, they additional honed this expertise to develop the interpretation options powered by on-device AI — which can be utilized for real-time translation throughout calls and throughout varied functions. Samsung Newsroom met with Yoonjung Choi and Yonghyun Ryu from Samsung Research’s Global AI Center to study extra about these bold options.
Smooth Communication and Strong Security With On-Device AI
On-device AI is the important thing differentiator to Galaxy AI’s Interpreter and Live Translate options — launched to customers by means of the discharge of the Galaxy S24 collection earlier this 12 months. Leveraging the superior computing assets constructed into the units themselves, smartphones with on-device AI can present providers with out counting on servers or the cloud. Users will be assured that their knowledge will stay non-public and safe since info shouldn’t be shared with exterior sources.
Samsung Research’s Global AI Center contributed to integrating proprietary expertise into its AI translation mannequin to commercialize these options for widespread use.
Envisioning a variety of functions, the Samsung Research workforce and MX R&D Office obtained an expansive quantity of knowledge. “We collected colloquial data for real-time translation during calls in Live Translate and travel-related data for Interpreter,” defined Yoonjung Choi, who led the mission. “To provide the most accurate translation, we studied and incorporated casual language used in chatrooms and HTML tags used in web browsers.”
How Samsung Research Trained Its AI Model
The Samsung Research workforce’s AI translation mannequin is predicated on deep studying expertise that learns from its personal knowledge. Yonghyun Ryu, who’s answerable for AI analysis and improvement, likened this course of to elevating a baby. “Similar to how a child needs excellent educational resources and caregivers to grow and thrive, good language data and talented researchers are required when developing a high-performance AI translation model,” he described.
Samsung Research has each — since 2013, the corporate has been offering in-house translation providers, conducting R&D associated to AI translation and accumulating high-quality knowledge.
Samsung Research’s workforce of deep studying specialists performed an essential position in coaching the AI mannequin. “If incorrect translations occur during the research and development process, it is necessary to identify the problem and make improvements. However, this can be challenging and time-consuming for researchers without sufficient capabilities and experience,” he defined. “Our researchers used their expertise and know-how to quickly analyze the cause of the issue and come up with a solution to enhance the AI translation model.”
To assess the efficiency of Galaxy AI’s translations, the Samsung Research workforce used quantitative metrics primarily based on take a look at units in addition to qualitative evaluations by human translators and the MX R&D Office.
In addition, the workforce gained credibility by competing in international machine translation competitions. “Although participation requires time and effort, good performance in competitions provides momentum for research and development,” Ryu…