The global interest in artificial intelligence (AI) has surged due to its capacity to enhance various aspects of people’s lives, providing convenience through multiple tools and devices. AI can efficiently convert sounds into statements, statements into images, or even transform statements into audible content.
With the advancement of AI, Dr Pongsatorn Janyoi, who teaches at the College of Computing, Khon Kaen University, has recently embarked on creating and implementing Wow Ja’s AI technology in his research. Wow Ja’ AI is an innovative technology aiming to transform statements into the Isan spoken language, a dialect predominantly spoken in the northeastern region of Thailand.
At the present time, Dr Pongsatorn finds that some Isan children need to improve in speaking Isan. Preserving the local dialect poses a challenge as children are exposed to diverse cultures, potentially overwhelming them with various influences. In response to this cultural challenge, Dr Pongsatorn has developed technology that can significantly preserve the Isan dialect.
The idea of preserving the Isan dialect through technology has inspired him to embark on a dedicated study and research journey focused on Natural Language and Speech Processing (NLSP), a branch of artificial intelligence that aims to enable computers to comprehend and process human language, particularly transforming statements into spoken Isan language.
During his master’s and PhD studies, under the supervision of Asst Prof Pussadee Sirisaengtrakoon, he collected over 5,000 sentences of the central Isan language spoken by native speakers. This extensive audio record was processed statistically, successfully developing the first version of sound transformation using AI.
Later, Dr Pongsatorn further will improve the technology with the invention of the second version, Wow Ja, which utilised Deep Learning techniques to transform statements into accurate Isan sound. The addition of a female voice was also incorporated into the system.
Through his efforts, Dr Pongsatorn’s work has yielded fruitful outcomes that facilitate the preservation of the local dialect and serve as a foundation for cultural conservation. By utilising technology, his research has successfully bridged the gap between technology and language, allowing for the transformation of statements into spoken Isan sound. This groundbreaking achievement contributes to preserving the unique Isan cultural heritage, ensuring its continued existence for future generations.
While many AI applications cater to communication needs in English and Thai, the introduction of Wow Ja’ AI demonstrates a remarkable integration of technology with the Isan language, catering specifically to the local community and their unique linguistic requirements.
With the user-friendly interface of Wow Ja, transforming written Isan text into spoken Isan sound is simple. The AI system synthesises the text and generates the corresponding Isan sound by typing the Isan reading. This versatile technology finds applications in various contexts.
At present, Wow Ja AI initially took place on the Khon Kaen City Bus, a public transportation service, where it is utilised to announce the different stopping points along the bus routes. This technological advancement not only serves as a means of improving communication but also plays a significant role in preserving the local dialect and fostering the cultural identity of the Isan region. Additionally, it has become a distinctive feature that leaves a lasting impression on tourists and passengers while travelling in Thailand.