We publish the second part of the review about the videoconferencing market. What developments have appeared over the past year, how do they penetrate our lives and become familiar. Above is a screenshot of the SRI International video, you can watch it towards the end of the article.
:
— Videoconferencing market — global cross-section
— Hardware vs software video communication
- Huddle rooms - aquariums
— Who wins: mergers and acquisitions
- Not video single
— Competition or integration?
- Compression and data transfer
Part 2:
—
—
Smart conferences
The field of videoconferencing is quite mobile in terms of introducing new technologies, many developments appear every year. Machine learning and artificial intelligence significantly expand the possibilities.
The speech-to-text technology has become the closest to reality and in demand. The machine recognizes a clear articulate speech quite successfully, but so far it is not very good with parsing by voices. However, video communication simplifies the procedure with consecutive replicas over different channels, and speech recognition-based services have already been announced by many vendors.
In addition to live captioning, which is convenient for people who are hard of hearing or in public places, businesses also need tools to manage the outcome of meetings. Tons of videos are inconvenient to review, someone has to keep minutes, fix agreements, turn them into plans. So far, a person helps to mark up and sort the decrypted text, but this is already much more convenient than writing it down in a notebook yourself. If necessary, it is much easier to search after the fact in deciphered texts and created tags. Integration with schedulers and various project management services significantly increases the efficiency of video communications. For example, Microsoft, BlueJeans are working in this direction. Cisco bought Voicea for this purpose.
Of the popular features, it is worth noting the replacement of the background. Any image can be placed behind the back of the speaker. This feature has been available to various manufacturers, including the Russian TrueConf, for a long time. Previously, for its implementation, a chromakey (green banner or wall) was needed behind the speaker. Now there are already solutions that can do without it - for example, Zoom. Literally on the eve of the release of the material, the replacement of the background was announced in Microsoft Teams.
Microsoft also knows how to make people transparent. In August 2019, Teams Rooms introduced the Intelligent Capture feature. In addition to the main camera, which is designed to capture people, an additional content camera is also used, the task of which is to broadcast an image of a conventional marker board on which the speaker can write or draw something. If the speaker gets carried away and blocks what he has written, the system will make it translucent and restore the image from the content camera.
Intelligent Capture, Microsoft
Agora made an emotion recognition algorithm. A system based on a cloud server processes video data, highlights faces on them and informs the user about what emotions the interlocutor is currently demonstrating. With an indication of the degree of accuracy of the determination. So far, the solution only works for one-on-one communication, but in the future it is planned to implement this for multi-user conferences. The product is based on deep learning, in particular, the Keras and TensorFlow libraries are used.
Emotion recognition from Agora
A fundamentally new field of application for video conferencing systems has been opened by technology that understands sign language. The GnoSys application was created by Evalk from the Netherlands. The service recognizes all popular sign languages. All you have to do is place your phone or tablet in front of you during a video call or regular conversation. GnoSys will translate from sign language and play your speech to the interlocutor sitting opposite or on the other side of the screen. Information about the development of Evalk appeared in February 2019. Then the project was partnered by the Indian Association of the Hearing Impaired - the National Deaf Association. Thanks to her assistance, developers gained access to a huge amount of data on sign languages, dialects and nuances of use, and there was active testing in India.
Now the issue of leakage of confidential information from negotiations is becoming very relevant. Zoom at the beginning of 2019 announced the introduction of an ultrasonic signature. Each video is supplied with a special ultrasonic code that allows you to track the source of information leakage in the event that the recording gets on the Internet.
Virtual and augmented reality are also making their way into the realm of video conferencing. Microsoft is offering to use the new HoloLens 2 glasses with its Teams cloud collaboration service.
HoloLens 2, Microsoft
Belgian startup Mimesys has gone even further. The company has developed a virtual presence technology that allows you to create a model of a person (avatar) and place it in a common workspace, which can be observed using virtual reality glasses. Mimesys has been acquired by Magic Leap, a globally renowned manufacturer of VR glasses. Industry experts strongly link the development of virtual and augmented reality technologies with the development of 5G mobile networks, since only they can provide the necessary speed and reliability to make such services available to a wide range of customers.
Collaboration on a project in virtual reality, photo Mimesys
Unusual cases. Robot control and law enforcement
In conclusion, a little about how the scope of video communications is expanding. The most obvious is the remote control of mechanisms in a dangerous area and uncomfortable environments, ridding people of dangerous or routine work. In the news field over the past year, management topics have been encountered, for example, , , . Solutions for the penitentiary and law enforcement systems are being developed.
So recently there was information about a new development of the research institute SRI International (USA), where the problem of police safety is quite acute. According to statistics, every year about 4,5 thousand attacks are carried out by aggressive drivers on law enforcement officers. Approximately one hundredth of these cases ends in the death of a police officer.
The development is a complex system that is mounted on a patrol car. It is equipped with high-definition cameras, display, speakers, and microphones. There is also a breathalyzer, a scanner for checking the authenticity of documents and a printer for issuing penalty receipts. Since the monitor of the complex is touch-sensitive, it can be used to conduct special tests to assess the general condition and adequacy of the driver. When the police crew stops the violator, the device extends towards the vehicle being checked and blocks its movement until all verification procedures are completed using a special studded bar at the level of the wheels. The system is already undergoing final testing.
Prisons have become another environment for the use of videoconferencing. Several US penitentiaries in Missouri, Indiana, and Mississippi have replaced regular short visits for inmates with video communication.
Communication through the videoconferencing terminal in one of the US prisons, photo by Natasha Haverty, nhpr.org
Prisons thus not only increase security, but also reduce costs. Indeed, in order to deliver a prisoner to the visiting room and back, it is necessary to provide a whole range of security measures along the entire route and during communication. Since visits are allowed in US prisons once a week, for large facilities with a large contingent, this process is ensured almost continuously. If we replace face-to-face meetings with video calls, then there will be fewer potential problems, and the staff of escorts can be reduced.
Human rights activists and the prisoners themselves say that in the current version, the video communication system significantly loses to personal communication and is in no way equivalent, even despite the increased talk time. Relatives do not have to go to prison, communication can be conducted from home, but in this case the cost of communication is much more expensive - from a few tens of cents to ten US dollars per minute, depending on the region. You can chat through local terminals on the territory of the prison for free.
Prisons that have tried to implement such communication systems are very pleased with the result and do not plan to abandon this practice. Independent sources note that the administration may be interested in implementing the technology because of the commission from videoconferencing operators who install their solutions there. In all cases, we are talking about special closed systems, the quality of which, according to American journalists, is inferior to popular services like Skype.
The video conferencing market will continue to grow. This is especially evident now, in the midst of the epidemic. Going to the clouds has opened up opportunities that have not yet been fully realized, and new technologies are on the way. Videoconferencing is "getting smarter", integrating into the common business space and continues to improve.
We thank Igor Kirillov for preparing the material and the editors of V+K for updating it.
Source: habr.com