Rolaxit Innovation Ltd.

We bring future today!
RolaxIT Innovation is a unique combination of experts, from seniors with a remarkable past in mathematics and invention, to young talents in the field of Artificial Intelligence and neural network programming.
The company was launched in early 2020 and its main activity is the development and implementation of systems based on AI Technologies.
The founder of the company is George Rusu.
George Rusu graduated from the Faculty of Mathematics and Physics of the University of Bucharest in 1965, the Faculty of International Economic Relations of the Bucharest Academy of Economic Sciences in 1979 and has a degree in Cyber Systems Technology from UNESCO. He holds a PhD in Probability and Mathematical Statistics.
He was part of the team of programmers for the first Romanian CIFA-3 computer from the University of Bucharest.
Gheorghe Rusu participated in the elaboration and implementation of important projects realized in Romania in the eighties: atomic industry, People’s House, Bucharest Metro, Refineries and important factories of the Romanian economy, construction and assembly companies, electrical networks, export of information technology to Algeria, Syria, Iraq, Egypt, Iran, etc.
After 1990, Gheorghe Rusu founded and participated in several companies in the field of informatics, telecommunications and technological process management for the introduction of the latest concepts: local LAN networks and national networks, complex SUN computer stations, fibre optic networks, the first network of commercial internet services, VOIP services, call centre, wireless networks, cable television, satellite, internet etc.
In 2001, Gheorghe Rusu initiated and was a founding member of the National Association of Internet Service Providers in Romania (ANISP), of which he was the first president. Through the software companies in which he was a shareholder Gheorghe Rusu was a member and President of ARIES and a member of ATIC.
Since 2013 Gheorghe Rusu has settled in the UK and attended the postgraduate course, Machine Learning, at Stanford University, San Francisco. After graduation, he initiated, developed and assisted the implementation of Artificial Intelligence projects in Spain, UK, Taiwan and USA. He is also a mentor on IA Alexa projects, Amazon product, for UK companies.
The team’s competencies cover the areas: web, chatbot, voice-bot, speech to text, text to speech, translation, Artificial Intelligence and Machine Learning. We are part of an international ecosystem, and we have partnerships with the most relevant names in the IT industry.
It is based on numerous in-house developed Deep Learning and Machine Learning algorithms for Speech-to-Text (S2T), Natural Language Processing (NLP), Natural Language Understanding (NLU), and Text-to-Speech (T2S).
At the same time, programming languages and training tools are used for the realized applications.

The services provided:
1. Transcription of voice into text
The people who do the transcript work audio to text have to concentrate many hours, a
hard and difficult job. The must carefully follows the recorded file and try to reproduce in writing what they hear as accurately as possible. It is a tough job with a lot of unpredictability. At the same time, their work must be efficient and respect the imposed deadlines.
With the development of information technology, the problem of transcribing audio into
texts using computers arose.
As is known, the recording of sounds, including speech, was a problem solved over 100 years ago. So, the problem of recognizing the sounds that make up human speech had to be solved. That’s how the interest in speech to text appeared.
Computing and artificial intelligence are behind the advances in this space. With massive amounts of speech data combined with faster processing, speech recognition has hit a point where its capabilities are roughly on par with humans.
How Do Transcript Audio Work?
Audio is stored in the form of electromagnetic signals. If you need to outsource audio transcription, then there are two main types of service providers: the human and the automated.
Human transcription combines automatic transcription and the work of highly experienced human transcribers. Transcription accuracy reaches 99-100%. Voice-to-text transcription time is reduced by more than 50% compared to using only human transcription. The service fee depends on several factors: the quality of the recording, the number of participants in the discussions, the language of the recording, etc. Human transcription is necessary in fields with field-specific terminology: medical. legal, technological, etc.
Automatic transcription is a service through which the user takes the transcribed text and edits it as he sees fit. Voice-to-text transcription time is reduced by more than 90% compared to using human transcription. Transcription accuracy reaches 90-95%. It largely depends on the quality of the audio recording
If the audio recording has sounds that are difficult to transcribe – in other words, poor audio quality, sound with distracting background noise, multiple speakers, crosstalk, speakers with unclear accents, speakers who speak quietly, or speakers who use a lot of jargon that computers can use I don’t understand, the accuracy of automatic transcription is low. By mixing automatic transcription with professional transcribers, the accuracy increases to 100%.
The speed with which the computer executes transcription is a basic argument in the use and development of automatic transcription services. In this way, people get rid of a tedious and repetitive work.
Fortunately, there is software available to help with these difficult processes. The software usually starts at phenomenal speeds when they want to do this kind of work.
The automatically made transcription is edited by professional transcriptionists who bring the transcription to a high level of accuracy and clarity.
In this way, the time needed to transcribe voice into text is greatly shortened and allows people to do a much easier job and with good results.

2 Translate audio into text
To translate an audio file and obtain a text file in a different language than the original one, two steps must be completed: transcribing the audio file into the text feeder and after editing the text, it is translated into the target language. Of course, the translation can be done by a human translator from the original language to the target language without going through the transcription phase. If the goal is to translate an audio file into another audio file in the target language, a human translator can perform the work.
The work would be quite hard and tiring. The duration would be approx. 7-8 hours for a one-hour recording. The costs are reasonable. For example, for English to Spanish translation, the rates are around 150-200 dollars per recording hour.
By using automatic transcription and translation, time and, respectively, costs are reduced by 60-90%.
This type of translation may be necessary in cases such as these:
• The translation of a journalistic audio interview into text
• The translation of an oral meeting report into text
• The audio to text translation of a conference, or a symposium
• The audio to text translation of an individual or group interview
• The audio to text translation of a hearing, a trial, or a legal procedure
• The audio translation into text of a sociological, medical, or scientific interview
• The audio to text translation of a podcast, a phone call, or a video conference.

3 Mobile Transcription App Features
Transcription of audio to text is the application made on desktop, laptop, tablet, mobile.
Fast transcription for students, researchers, professionals, legal groups, medical practitioners, and more is done on mobile. Account creation for application users and editing is done on computers and mobile application use, registration, sending to transcription and the transcribed file.

Who is the audio to text transcription for?

Transcribers work in a wide range of industries including legal, medical, conference, and educational institutions. Transcription is used by millions of people for meetings, interviews, lectures, and other conversations. The transcribed can be searchable, copy and paste, or rendered as text-based content.
People transcribe audio in researching human language, court proceedings, or in marketing campaigns and PR activity.

The next stage

According to the study, the U.S. transcription market size was valued at USD 19.8 billion in 2019 and is anticipated to expand at a CAGR of 6.1% from 2020 to 2027. Organizations across the globe generate large volumes of data every day that can be effectively used for obtaining valuable insights. However, data accuracy is of paramount importance if these data-driven insights are to be used for making business decisions. Owing to this, organizations have started opting for transcription services so that critical information can be recorded accurately.
The trend of automatic speech recognition is gaining traction, which is expected to propel the growth of the transcription market during the forecast period. The market is projected to be driven by several factors such as growing demand for automated transcription services, organizations to leverage transcription services for market
research, timely delivery, and improved quality of service offered through transcription services.