It’s a kind of magic ✨

Our team of professional linguists has prepared a short video for you about the basic principles of automatic speech recognition (ASR).

In the video you will learn what a phone is, what takes place in the acoustic and linguistic model in ASR, and how spoken language becomes written words.

This process is the basis of our technologies for automatic speech-to-text conversion, which we have developed together with the Technical University of Liberec. Believe it or not, even for us it’s still a kind of magic. 🔮🧙

The video is in English and comes with English or Czech subtitles.


Střih: Kateřina Morozová

About the author:

Lenka Weingartová is a linguist, phonetician and logician by education, she even has a doctorate in phonetics. After working on the development of synthesis of the Czech language for Google, she moved to Newton Technologies in 2016, where she took over the development and maintenance of so-called ‘dictionaries’ – language models for speech recognition. In recent years, she has also been doing research on the borderline between recognition and NLP (computer language processing). In her free time, she sings in a band and in a choir, enjoys reading science fiction and fantasy, playing computer games, resolving ciphers and traveling the world. It is a purebred owl, so you won’t spot her in the office until the afternoon.