Developing Multi Translation Chat Application Using Django Frameworks and M2m100 Model

Abstract

The significance of language as a powerful communication tool and its evolution in the context of modern communication platforms. While chat applications have provided convenient means for global interactions, language barriers and misinterpretations often arise due to the difficulty in capturing the intended meaning. Although translation technologies like Google, Yandex, and Bing exist, constantly switching between chat and translation applications can hamper user productivity. To address this limitation, some chat applications integrate translation functions, enabling users to translate texts without leaving the application. However, these functions have certain limitations, such as the need to select and translate specific chats and the inability to translate emoticons or stickers. To overcome these limitations, the development of a web-based chat application capable of performing direct translation. By using the utilization of the M2M100 NLP model in chat application development. The M2M100 model stands out by offering translation capabilities for a wide range of languages, surpassing other translation APIs such as those from Google and Microsoft which require subscription costs. Being an open-source model, M2M100 proves to be a favorable choice for initiating the development of the chat application. Despite the longer process involved, the model's results are deemed satisfactory