Accessibility Enhancement for Air Traffic Control

The aim of the project is to provide a speech to text solution for air traffic control conversations particularly pilot – controller communication.
The basic step is to extend speech recognition models for decoding ATC specific commands and phrases.
Enhance the ATC speech signal by de-noising for improved speech recognition accuracy.

Automatic Speech Recognition (ASR) in a nutshell:

Our approach:

Initial results:

Future work includes preparing noise adulterated data to match ATC conditions (due to lack of available open source data) and training KALDI models for an improved accuracy.

People: Harshpal Singh