NEWTON SpeechGrid
NEWTON SpeechGrid
A comprehensive solution for the automatic transcription and processing of audio recordings
A comprehensive solution for the automatic transcription and processing of audio recordings
Try it out Features
What is NEWTON SpeechGrid
NEWTON SpeechGrid is a complete workflow for the automatic transcription and processing of audio recordings. It includes NEWTON NanoGrid, a powerful Cloud infrastructure with APIs for online speech recognition. The version NEWTON MiniGrid is also available and is recommended for use in closed business systems. Demo projects are also included.
Trying out NEWTON SpeechGrid is easy thanks to the test application.
Key features
Fast access to highly accurately transcribed audio recordings of interviews, meetings, podcasts and other sources. Our technologies can recognize and transcribe speech in 18 languages.
Performance
High performance, speed of recognition and guaranteed availability thanks to the scalable infrastructure.
Security
For applications with high security demands we offer an isolated on-premise solution.
Dictionaries
Highly accurate speech recognition thanks to the extensive dictionaries, with the option of adding user words, for example specialized terminology.
We understand the voice of 1 billion people
Our applications can automatically recognize speech with high accuracy for all of the major Slavic languages. Furthermore, we specialize in languages spoken in regions where our company operates: Central, Eastern and Western Europe.

Bosnian
Bulgarian
Czech
Croatian
English
Hungarian
Macedonian
Montenegrin
Polish
Romanian
Russian
Serbian
Slovak
Slovenian
Spanish
Ukrainian
Specialized dictionaries
For the best recognition results we work with specialized dictionaries and acoustic models. The availability of dictionaries may vary from language to language.
Offline transcription
Universal dictionary designed for the transcription of audio recordings.
For dictation
A dictionary containing general vocabulary for direct dictation and instant conversion of voice to text is offered.
Telephone dictionary
Specialized dictionary and acoustic model designed for transcribing telephone calls.
Interested in using Nanogrid speech recognition in your application?
See our APIs and projects available on GitLab, which describe the possibilities of connecting to Nanogrid.
Our APIs
- REST API
- Websocket
- GRPC
Detailed description NanoGrid API.
If you add your user words into the Nanogrid lexicon, you need to use the supported phonetic alphabet. Learn more about how to proceed here: https://nanotrix-models.gitlab.io/g2p-manuals/
Demo projects
.NET Starter
Demo project for .NET developers, which demonstrates the integration possibilities.
Projekt: https://gitlab.com/nanotrix/ntx-net-starter
Ntx-JS
Simple web REST API or websocket API. Available with demo for Node.js or CURL 🙂
Projekt: https://gitlab.com/nanotrix/ntx-js
Ntx
GRPC client in .NET Core.
Projekt: https://gitlab.com/nanotrix/ntx