OpenPaaS’ Newsletter — September 2018
Aiming for the star
Exploring uncharted territories
Just like a small solar system, LINAGORA is made out of multiple independent entities that are all revolving around the sun. For us, those astronomical objects represent products such as OpenPaaS or LinShare, or company offices located in a handful of countries including France, Vietnam, Canada and Tunisia.
In this analogy, the sun illustrates the sense of purpose within the company, perhaps most importantly the support for free software — for which we all thrive — and which is illuminating the path so that we can go further, both individually and collectively.
Just as in outer space with the Voyager space probes, explorers are needed when it comes to discovering new planets, building new systems and helping us to better comprehend reality.
Members of the Research and development (R&D) team, which are part of the LINAGORA Lab, are not unlike those explorers. Their research is used to nurture in-house software products like OpenPaaS, our open source collaborative platform.
This newsletter will be dedicated to showcase some of their most recent results, and will give us a glimpse into what is coming next for OpenPaaS when it comes to email processing.
Research fuels development
The LINAGORA Lab, based in Toulouse, is working closely with universities, contributing to lead researches on topics ranging from cloud computing to voice processing. Within LINAGORA, those innovations are used to propel our products.
Take OpenPaaS for instance. As explained by Michael Bailly during his interview, OpenPaaS started as a research project, originally called OpenPaaS:NG, for “New Generation”. This research project has contributed to laid the foundations for what would later become the integrated platform of today.
More recently, in April this year, LinTO has officially started as a research project: the new generation of open source smart assistant will receive three years worth of public funding.
If you wish to learn more about LinTO, please have a look at this previous article (also, I have heard that another article was planned on that matter. Stay tuned on the company-wide social account to stay posted).
Mail overload as a modern-day society problem
Answering emails takes a toll on productivity. Often, just a simple answer could suffice, but writing it might become tedious overtime, especially if the process has to be repeated more than a dozen time per day, such as when you seek the best time for an appointment.
That is a good news… at least for software engineers, as they are always looking for clever ways to automate repetitive tasks. How can we use intent detection to select emails that require a reply and pre-write a pertinent answer for those? Manon, which recently worked for LINAGORA Labs to complete her master thesis in linguistics, took part to the development of a OpenPaaS feature to do that. The approach, the tool and some interesting results are presented in the following medium article.
The main takeaway?
Our tool is able to analyze incoming emails and decide which one requires a response and then retrieve relevant items to be able to generate an adequate answer.
What do they would like to do in the future?
We are planning to link the answering mechanism to the OpenPaaS calendar to automatically check availability and generate answers depending on the user’s constraints. We also plan to detect and answer to other intents as administrative workflows […]
How can computer better recognize the human voice?
How can we, as humans, recognize a voice in a noisy environment? How is our brain able to tell apart background noises from significant and meaningful sounds? More importantly for us, how can we transfer this ability to computers?
Voice processing is one of the problem computational linguistics tries to tackle. But before voice can be processed, it needs to be detected. Voice activity detection (VAD) represents the ability for a computer to detect the human voice among other noises. How can a machine do the same?
Please read article by Rudy Baraglia to learn more about it, starting from the basics.
Conclusion
We hope you have enjoyed reading this newsletter. If you do, do not hesitate to leave a comment in the section bellow.
Please keep in touch with OpenPaaS on Twitter, Facebook, GitHub.
Interested in joining LINAGORA? We are hiring!
