TL;DR:
- Swindon Council in the UK has successfully implemented a machine learning solution to reduce translation costs and turnaround time.
- The solution cut content translation costs from £160 per document to just 7p and reduced the time from weeks to minutes.
- The machine learning tool, developed in collaboration with Amazon Web Services (AWS), streamlined the translation process and ensured security and user-friendliness.
- Swindon Council’s initiative has led to significant cost savings and increased efficiency, with translation spending reduced from £64,000 to £27 for the Paediatric Therapy Service.
- The developed tool has been made available as open source via GitHub and AWS Industry Solutions, benefiting other local governments.
- The solution is now being adopted by other UK councils and international organizations, including Edinburgh, France, and Spain.
Main AI News:
In a groundbreaking move, Swindon Council in the UK has successfully implemented a machine learning solution that has significantly reduced content translation costs and turnaround time. The council, with a population of 230,000 and over 100 languages spoken across the borough, faced the challenge of translating approximately 400 documents annually for its Paediatric Therapy Service. These translations, which previously cost an average of £160 each and took up to 16 days, were both time-consuming and costly.
To address this issue, Swindon’s Emerging Technologies team undertook a comprehensive market assessment to explore solutions that would streamline the translation process while keeping costs to a minimum. After careful consideration, they selected Amazon Web Services (AWS), leveraging the company’s existing technology for a unique application.
Working closely with AWS, the team developed a tailor-made machine learning tool based on the Amazon Translate neural machine translation service. The objective was to create a system that was not only fast and user-friendly but also secure, ensuring the protection of sensitive child information. The result was a solution that significantly simplified the translation process, requiring only three clicks to complete. Moreover, the team designed an intuitive user interface, ensuring accessibility for all users.
The impact of this innovative solution has been remarkable. With no upfront costs apart from staff time, Swindon Council has seen translation spending for the Paediatric Therapy Service decrease from £64,000 annually to a mere £27. Furthermore, the translation turnaround time has been reduced from weeks to an average of just 14 minutes per document. These impressive results have prompted the council to extend the use of the document translation tool to other departments, fostering greater efficiency and cost savings across the organization.
Swindon’s success story has not gone unnoticed. The council’s commitment to open source and knowledge sharing has led them to make the developed tool available to other local governments. As part of their contract with AWS, Swindon Council stipulated that the solution should be provided to other customers “at cost.” The tool can now be accessed via GitHub and AWS Industry Solutions, with translation services offered on a pay-as-you-go basis.
The impact of this initiative is spreading beyond the borders of the UK. AWS reports that the solution has already been adopted by other UK councils, including Edinburgh, as well as in countries like France and Spain. The value and potential of this innovation are recognized globally, with organizations worldwide set to benefit from the cost savings, improved efficiency, and enhanced accuracy offered by the Swindon Council-developed machine learning tool.
Swindon Council’s Emerging Technologies team continues to explore opportunities for technological advancements across various council services. Future endeavors include utilizing similar machine learning solutions to meet easy read requirements in adult social care. By partnering with like-minded organizations and sharing use cases, Swindon Council aims to spearhead a collaborative effort that brings substantial benefits to customers and propels the entire sector forward. As Swindon Council Chief Digital Officer Philip Murkin aptly puts it, “Together, we can achieve so much more and create a better future for all.”
Conclusion:
Swindon Council’s successful implementation of the machine learning solution for translation has far-reaching implications for the market. It showcases the potential of artificial intelligence in optimizing processes and reducing costs for public sector organizations. The significant cost savings achieved by Swindon Council demonstrate the viability of machine learning technology in tackling language barriers and improving efficiency in document translation. With the tool now available as open source, it presents a valuable opportunity for other local governments to replicate Swindon’s success. This development underscores the importance of collaboration, knowledge sharing, and open source initiatives in driving innovation and progress in the business and public sectors. The adoption of this solution by other UK councils and international entities further emphasizes the market demand for efficient and cost-effective translation solutions powered by machine learning.
