Access through a voice recognition system

Nov 24, 2022

Onvego, marketed by Dogma, is a system that is integrated into the development process of a voice recognition application and enables accessibility for websites and mobile and web applications for users with various disabilities (for example, the blind, visually impaired, and those with motor disabilities).

Onvego, marketed by Dogma, is a system that is integrated into the development process of a voice recognition application and enables accessibility for websites and mobile and web applications for users with various disabilities (for example, the blind, visually impaired, and those with motor disabilities).

The Onvego enables the conversion of the recorded voice into written texts through which various commands will occur for various systems/websites/applications

Such as searching for a vacation in the search engines, uploading a post to Twitter or Facebook, sending an SMS, sending an email, ordering a movie online, etc.

 

In the last two years, awareness has been growing in the areas of website accessibility, mobile and web app accessibility, and therefore we launched voice recognition using

Onvego's system (marketed through Dogma) which performs voice recognition and voice understanding of about 95%.

That is, we were not satisfied only with voice recognition, but also with the aspect of the ability of your application or website to understand what is being said,

Even if the language is Hebrew, even if the language is foreign and with a distorted accent, and even if the user phrased his words in an unclear manner.

 

For example: when we say the word cat - a voice recognition system may recognize it as Cut.

The Dragon system does not know how to recognize the correct word that the user meant. Therefore, Onvego added an algorithm that allows for the identification of an accent mark and the distinction of what the user meant to say in the context of the entire sentence.


Accessibility through voice recognition how we make it simple

Prof. Ami Moyal, who heads the Center for Language Processing at Afka College says:

"Speech recognition technology depends on the language. Every time we want to offer a complete solution for a new language, we will have to recruitat least a thousand people to be speakers, and only then will we be able to train the system.

This is a very expensive process that takes a lot of time, so most people only start with popular languages. Hebrew, naturally, entered thepicture very late or usually does not enter at all."

______________________________

The accuracy of the words read:Dragon's technology recognizes about 90% of the words read to it. In order to reach 90% accuracy it is important that the speech be slow in a quiet area and without background noise. This is a limitation that does not allow users to enjoy the technology simply and easily. Therefore, when you need to find a technological solution that will enable accessibility or the development of an application with a voice recognition system, consider that it is desirable to use a company that specializes in the development of voice recognition systems and access to websites and applications.

User interface operation UXThe Onvego user interface from Dogma is very convenient and easy to use and implement by application developers. You can record or type text and yes, you can easily 'train' the system to understand sentences in different contexts.

Options after recording:

  • cutting

  • copying

  • Sending the text by email and SMS

  • Language definition

  • Speech interruption detection

  • Setting up name recognition and more...


https://www.dogma.co.il/%D7%94%D7%A0%D7%92%D7%A9%D7%94-%D7%91%D7%90%D7%9E%D7%A6%D7%A2%D7%95%D7%AA-%D7%9E%D7%A2%D7%A8%D7%9B%D7%AA-%D7%96%D7%99%D7%94%D7%95%D7%99-%D7%A7%D7%95%D7%9C%D7%99/