Chair: Rian Schwarz-van Poppel
Chair: Dr. Carlo Eugeni
In this first talk in the Training and Competencies section, Tibor Varga and Katalin Varga analyzed the present state of typing education in Hungary. Hungary boasts an elaborate and well developed system of typing education, as it is both part of primary school and secondary school curricula, and taught at university level as well. In addition, there is a relatively large amount of typing education literature, software and tutorial and examination programs. In order to avoid future health problems, Tibor Varga stresses the importance of teaching children adequate working postures for computer work, like touch typing, at an early age. In this, two challenges need to be addressed: at first, children nowadays are very keen computer users and usually they have already acquired inadequate working postures before they are confronted with formal typing education. A second challenge for typing educators is a lack of motivation among the youngsters for their teachings.
Report by Michiel Haanen Historically, Intersteno competitions are based on languages that use alphabetical scripture. Consequently, until recently it was virtually impossible for people from countries where they use non-alphabetical scripture, such as Japan, to enter the competition. Takahiro Sumino has developed a method that makes it possible to implement Japanese scripture in Intersteno competitions. One of the puzzles Takahiro Sumino had to solve was: how many strokes does it take to produce Japanese characters? To fairly and accurately assess the typing skills of participants, he had to develop a conversion rate system. Thanks to Takahiro Sumino’s efforts, this year Japanese people were able to enter the Intersteno competition and compare their skills to those of people from countries where alphabetical scripture is used. Furthermore, people from countries with alphabetical scripture systems can now also test their skills on Japanese texts in Intersteno typing competitions. Sumino’s method was successfully implemented in the 2015 internet contest. In the future he intends to make this possible for other non-alphabetical languages, such as Vietnamese and Korean.
SeokRyeon Sohn (South Korea) Report: Désirée de Jong In his presentation mister SeokRyeon Sohn, member of the Intersteno Education Committee, talks about current activities of stenographers in the fields of public prosecution and education. In his country the process of prosecution has changed a few years ago. Nowadays the entire interrogation process of suspects and witnesses is recorded on video. Stenographers assist in the interrogation process by writing the minutes of the interrogation. The detailed and elaborate stenographic reports help to better protect the human rights of both suspects and people who testify against them, speed up legal procedures and increase transparency. In education, special programs have been introduced at the universities to secure the educational rights of students with disabilities. Hearing impaired students are provided with real-time captions that are written on-the-spot by specially trained stenographers. The vision impaired are helped with real-time braille translations. In some cases sign language or captions are provided through online remote controlled systems. The captions are written by certified stenographers, for whom a special educational program was set up by the Korea Smart Steno Association.
Report by Deru Schelhaas Video streaming of lectures became popular in recent years, for example in Massive Open Online Courses (MOOC’s) offered by universities. Captions make content accessible to the hearing impaired. They can also help other viewers to better understand a difficult topic. Automatic speech recognition (ASR) is of service in this, professor Kawahara from Kyoto University in Japan convincingly showed in his talk. Hardware and software infrastructure are ready, at least for the Japanese language, but a few problems have yet to be resolved. Beside ASR’s current inaccuracy rate of 5%-10%, these problems are mostly related to editorial choices. How to present the subtitles on a screen, for instance, since there are no guidelines for captions in internet videos? And to which side should the balance tip: an edited rendition to serve readability or a 100% verbatim rendition, which may very well not offer the best reading experience? Since the techniques rapidly evolve and the public demand for proper subtitling only grows, answers to these questions will be found sooner rather than later.
Reporter: Henk-Jan Eras According to Tsuguo Kaneko an impressive 13 million people in Japan are late-deafened or hard of hearing. People that were born deaf or became deaf at a very early age in general prefer sign language. People who are late-deafened or hard of hearing prefer modes of written communication. In the early days paper and pen were used for this. In the sixties communication with an overhead projector was developed: summarized writing. However, because a person can only write so fast, writing by hand covers only 25% of spoken word. With the introduction of the PC and the innovation of keyboard systems suitable for the Japanese language, a maximum of 300 characters was reached. In Japan automatic speech recognition (ASR) and other forms of speech to text technology (STT) are highly developed. They are used in both the Supreme Court and the House of Representatives, for example. By way of this communication by writing, people who are completely or partially deaf can access reports on their smartphones and tablets. In addition, a development towards bilateral communication is gaining ground. Late-deafened people who do not understand sign language can communicate smoothly using ASR and Artificial Voice Synthesizer (AVS). This service includes voice-to-text and text-to-voice translation. Another clear illustration of how a machine can compensate the weak points of man.
Kim Hanwoo Reporter: Jasmin Geijteman Overcoming the challenges of real-time captioning for broadcasting in remote places is the theme of the presentation of Kim Hanwoo. In Korea all the major broadcasting companies are based in Seoul, Korea’s capital, where almost half of the population lives. They are connected via leased lines to Korea Steno, the national capturing service which is based in Seoul as well. The broadcasting companies send their audio-visual material over the leased lines and Korea Steno provides the subtitles. 100% of national television broadcasts is subtitled. There was a problem with connecting to the peripheral stations of the broadcasting companies, though. These stations are too away from Seoul to use the very expensive leased lines. When regular internet connections and regular compression methods are used, the delay can amount up to eight seconds. That causes mismatches between the images and the subtitles, which is confusing for the viewer. The challenge is overcoming that problem. In January 2015, Korea Steno started subtitling local broadcasts using the Portable GrayMap-format to compress the television signal. This way a regular internet connection can be used, with the delay reduced to only 0,5 seconds. This enables Korea Steno to caption broadcasts even in remote Korean cities without mismatches between the images and subtitles, and at very low costs.
Report by Michiel Haanen Richard Gatarski is the founder of the Swedish company WestreamU. The company provides live streamed web TV or web video for all kinds of occasions, such as the opening ceremony of a conference like Intersteno. At one point a client asked him if his video production could be subtitled in real-time. Being the businessman he is, Richard said: of course! Then began his search for a decent, quick, reliable and affordable way to deliver this service. In this search many techniques and methods were tested and used, among others a DOS-based tool. These days Richard Gatarski hires the services of Text on Top, a Dutch company specialized in real-time captioning. One of the “captioneers” is Wim Gerbecks. During the presentation he provides the audience with a continuous stream of on the spot produced captions, that are projected on the big screen behind the speaker. To achieve this fast stream of captions Wim Gerbecks uses a special stenographic keyboard; the Velotype Pro keyboard. For Richard’s company a special new software tool was developed, which makes it possible to virtually immediately mix the subtitles with the images, so no post-production is needed.
Reporter: Roberto La Rocca 25 years ago Patricia Graves started her own captioning company, initially because she wanted to be of assistance to the hearing impaired in judicial contexts. She was soon to discover that the need for qualitatively high standing real time text captioning was much larger and wider. For instance, there is also a large potential for real time text captioning as an aid for non-native speakers of English in various situations, and for memorization, storage and reference purposes. According to Patricia, quality in captioning involves technological, linguistic (readability vs. verbatim rendering) and professional issues. The focus should be on the actual operators, on formulating and enhancing professional competencies, on defining common grounds (standardization of the present diversity of standards in the US) and on certification. The involvement of the academic world will be required in order to determine actual consumer needs and the acceptability of services worldwide, and to meet various technical challenges.
Report by Deru Schelhaas In his talk, Domingo José Mazza, head of the publications office of the Argentinian Senate, explained in a nutshell how the Argentinian state is constituted. Besides that, he touched on a variety of subjects, such as the need for modernization of the public sector in Argentina, standardization and management of the main process of publication and editorial checking of new legislation in order to guarantee and improve its quality. Furthermore, José Mazza revealed that 23% of the employees in his office have some sort of disability. This number is high, especially when compared to the 3% in the whole of the Argentinian public service. He also stressed the importance of constant self-evaluation: how to get rid of obsolete habits and flow like water in periods of change? In that regard, in conclusion José Mazza shared a lesson that is almost impossible to counter: you are as strong as your weakest link.
Reporter: Wouter Zwijnenburg Mr. Eugeni gives a presentation of Par.el.on, Parlamento Elettronico Online, an open source platform for direct democracy, integrating solutions from twenty years of eDemocracy experiments all over the world. The project originates from the Italian Five Star Movement, the populist, anti-establishment party started by Beppe Grillo, but is no longer connected to that party. The project focuses on increasing the influence of the people on the decision making process and the content of laws. Universities and programmers are involved to increase the electronic possibilities for the public to collaborate with members of parliament and the government in the law making process. Par.el.on is an electronic platform that allows personal access to the decision making process. The access is personalized. When you sign up, you are asked what you want. There are many possibilities: you can do a proposal, you can read other proposals, you can vote on other proposals. The transparant platform keeps track of your comments and of the contributions of other people. One of the important goals of the project is to guarantee maximum accessability. Because of the difficulty of legislative language, participants are informed by keywords to clarify the meaning of words and terms. At every step it is easy to understand in which phase of a law making process you are involved. The system offers a clear insight in the process and informs the contributors about new proposals. More information on www.parelon.com
1887 - 2019 All Rights Reserved. Intersteno - International Federation for Information and Communication Processing