A training method for multilingual neural machine translation systems that is efficiently extendable to new languages and data modalities.
The Challenge
Training neural machine translation systems is challenging both in terms of data and computational resources. These factors become more critical in the multilingual setting, where several languages may be input or generated by the system. The most widespread strategy consists of training a single sequenceto-sequence system shared between all languages. This architecture allows knowledge transfer but it forces a dependency between languages. This dependency limits the ability of the system to efficiently extend to new languages or modalities. The whole system has to be retrained using data for all languages which implies variations in the overall performance.
The Technology
This technology based on the sequence-to-sequence architecture for neural machine translation defines how to efficiently train a multilingual system extendable to new languages and modalities while allowing knowledge transfer.
The process consists of two main steps. Firstly, joint training of languages specific encoder and decoders to a common language representation without parameter sharing. Secondly, incremental training of new languages and modalities to the system, including an additional module to mitigate the differences between speech and text representations.
Innovative aspects and advantages
Multilingual neural machine translation system that is able to:
• Converge to a common language representation without sharing parameters between languages.
• Extend to new languages by a fraction of the cost of previous methods.
• Extend to spoken language even to language pairs without specific training (i.e zero-shot translation).
Current stage of development
(Development stage or status) The technology has been explored and verified by a proof of concept system with positve results on text and speech translation. Partners to further develop the system and/or to establish commercial agreements along with technical cooperation are sought.
Applications and Target Market
This method could be applied to any machine translation solution available to increase the efficiency of their systems to extend to new languages or input modalities.
"The project leading to this application has received funding from the European Research Council (ERC) under the European Union's Horizon 2020 research and innovation program (grant agreement No. 947657)"
The Universitat Politècnica de Catalunya - BarcelonaTech is a public institution dedicated to higher education and research in the fields of engineering, architecture and science, which contributes its knowledge and expertise in order to increase scientific output, transfer its results to society and provide a network of scientific and technical state-of-the-art facilities and technology valorization services that place us at the leading edge of innovation and economic development.
The UPC has established itself as a driver of innovation and is the technology partner of choice for companies and organizations with which it develops projects and builds partnerships. A role borne out by the numerous agreements and research projects that have been set in motion by groups, organizations and laboratories; the creation of new technology-based companies; the generation and exploitation of patents, and the scientific and technical services UPC makes available to its environment in order to generate progress and employment.
The Technology Transfer Office (SGI) is responsible of Designing, coordinating and implementing research valorisation strategies, carrying out the protection policy of the research results, marketing these results through license contracts and designing and setting up the University's enterprise creation model in order to transfer the results of the research to the market, protect and commercialize these results, promote the culture of entrepreneurship and innovation, and create technology-based companies within the UPC environment.
Create your free account to connect with Universitat Politècnica de Catalunya - UPC and thousands of other innovative organizations and professionals worldwide
Send a request for information
to Universitat Politècnica de Catalunya - UPC
Technology Offers on Innoget are directly posted
and managed by its members as well as evaluation of requests for information. Innoget is the trusted open innovation and science network aimed at directly connect industry needs with professionals online.
Need help requesting additional information or have questions regarding this Technology Offer?
Contact Innoget support
