We stand with Ukraine

Computational linguistics for today and tomorrow

Expert.ai Team - 17 May 2016

An explosion of text data and the success of virtual assistant applications are making natural language understanding a must-have in consumer tech. In this post, we’ll take a closer look at computational linguistics, one of the key applications behind speech recognition and natural language processing technologies.

Applications like Siri and some of the newest Internet of Things-based intelligent systems with natural language interfaces depend on technologies that can help machines understand human language. Computational linguistics exists at the intersection of linguistics and computer science for building software and systems that are able to process, understand and produce language for various applications, whether for dialogue, or virtual assistant type functions using information retrieval from knowledge bases.

Computational Linguistics helps us interact with machines

Helping machines understand language or take action based on what is being said or written is at the heart of some of today’s biggest artificial intelligence challenges. To support the language understanding part of the equation, computational linguists are specialized in helping machines process natural language, the kind that we use when we speak and write. To do so, computational linguists develop formal models that simulate aspects of human language and systems that facilitate human-computer interaction. They also work to automate a range of activities such as machine translation, text summarization, speech recognition and generation, information extraction and retrieval, and sentiment analysis of text.

Computational Linguistics is changing our lives…

The latest advances in computational linguistics will continue to improve the usability of “smart” systems that allow us to communicate with computers using natural language. These are just a few of the ways that systems that are supported by computational linguistics are already changing how we communicate with people, companies and technology:

  • Rather than going through an exhaustive telephone menu system, many companies, especially banks, are offering natural language based customer service to handle simple queries (connecting with Q&A knowledge bases or how-to type information) that the customer can speak or write in his or her own language.
  • Digital personal assistants or other intelligent agents use a combination of approaches, including computational linguistics, to perform a variety of functions, from limited conversational dialogue, to responding to requests, typing out dictated text (such as for text messages or emails), or retrieving information, like Apple’s Siri.
  • Combined with artificial intelligence mechanisms, such systems can go one step further to actually take action upon our requests, to make restaurant reservations, set up meetings or arrange other services. Among other abilities such as recognizing moving objects, faces and gestures, Honda’s robot assistant, ASIMO, verbally responds to questions in different languages as part of its human-like interaction.
  • Internet of Things systems combined with voice-recognition technologies are being developed to complement home-based electronics systems for temperature control, internet, electricity, sound and more, creating a new market of high-end, smart-home devices. Amazon’s Echo is a version of one aspect of this, serving as a voice command device that can play audio (music, podcasts, audio books), provide real-time information on the weather, traffic, etc. or set alarms, or communicate with other smart devices.

… And will continue to change our future

While we still do not know exactly how people produce and comprehend language, thanks to computational linguistics, we’ll get there eventually. Until then, the future of computational linguistics is full of interesting applications that will continue to transform how we live and work.