Periodic Reporting for period 1 - MultiConvAI (Enabling Multilingual Conversational AI)
Reporting period: 2021-01-01 to 2022-06-30
In recent past, Conversational Artificial Intelligence (AI) has made major advances, thanks to the availability of big data and
increasingly powerful deep learning. Task-based statistical dialogue systems (SDS) are now viable, embedded in popular commercial
applications (e.g. the Apple’s Siri, Amazon’s Echo, Google’s Assistant) and cost-effective in many scenarios (e.g. customer support,
call centre service, searching, booking). Yet current SDSs are only available for a handful of resource-rich languages, leaving the
majority of the worlds languages and their speakers behind. Our project has developed the first prototype system for scaling
conversational AI to multiple languages. This is based on new methodology that learns multilingual representations
without the need for expensive training data, using a process called semantic specialisation that complements
representations with common-sense and linguistic knowledge in external knowledge graphs. Building on our promising pilot studies,
we have developed Natural Language Understanding (NLU) modules for SDS via using more effective semantic specialisation
based on joint multisource multi-target training and have focussed on typologically-diverse languages.
We have proposed new methods for selective sharing and structural adaptation and optimisation for the target languages
guided by typological knowledge. The best resulting technology was integrated in a demo prototype system which can generate multilingual
NLU input for more widely portable SDS. On a larger scale, extending the multilingual scope of SDSs can have major socioeconomic benefits:
it can broaden the global reach of conversational AI and it can enhance its commercial viability.
increasingly powerful deep learning. Task-based statistical dialogue systems (SDS) are now viable, embedded in popular commercial
applications (e.g. the Apple’s Siri, Amazon’s Echo, Google’s Assistant) and cost-effective in many scenarios (e.g. customer support,
call centre service, searching, booking). Yet current SDSs are only available for a handful of resource-rich languages, leaving the
majority of the worlds languages and their speakers behind. Our project has developed the first prototype system for scaling
conversational AI to multiple languages. This is based on new methodology that learns multilingual representations
without the need for expensive training data, using a process called semantic specialisation that complements
representations with common-sense and linguistic knowledge in external knowledge graphs. Building on our promising pilot studies,
we have developed Natural Language Understanding (NLU) modules for SDS via using more effective semantic specialisation
based on joint multisource multi-target training and have focussed on typologically-diverse languages.
We have proposed new methods for selective sharing and structural adaptation and optimisation for the target languages
guided by typological knowledge. The best resulting technology was integrated in a demo prototype system which can generate multilingual
NLU input for more widely portable SDS. On a larger scale, extending the multilingual scope of SDSs can have major socioeconomic benefits:
it can broaden the global reach of conversational AI and it can enhance its commercial viability.