With our current speech-to-text/text-to-speech technology and conversational agents like Siri and Alexa seemingly everywhere, why aren't we creating powerful new tools to support language learning? What is holding us back? What do we need in order to push forward?
This language exists ... it's the natural human language :) the problem is that, if we want to implement such natural language, it depends on the context, the semantics is very unstable and chaotic (look at the current and old artistic expressions ..) also depends on the ontology of the domain. Definition of a language with the Mantague semantics of natural languages is possible ... there is even some realization in language prolog ... certainly, we may engages this type of research .... but it is one of the General Problem Solver's limitations of artificial intelligence. the work requires a long way ...
You can get reasonable results if the chatbot is domain specific, and you have already built a specific ontology and machine learning algorithm there (such as a bidirectional LSTM)
Vikas Ramachandra,
Thanks for the response.
Can you give an example of such a chatbot/conversational agent that is accessible online?
Thanks!
there is a list here, (just google search and you fill find several, you can evaluate them for your specific application)
http://www.businessinsider.com/best-ai-chatbots-online-robot-chat-2017-10
Thanks.
I am familiar with that list; none of those chatbots quite make it as a language learning partner.
The question remains, what precisely is holding us back in this area?
Is it just a lack of will on the part of developers or are we just not there yet--in terms of technology?
- Badges
- Science topic
- Similar topics
- Computer Science and Engineering
- Artificial Intelligence