Call-sign recognition and understanding for noisy air-traffic transcripts using surveillance information

TL;DR

This paper presents a novel call-sign recognition system for noisy air traffic communication that leverages surveillance data to significantly improve accuracy, especially in low SNR conditions.

Contribution

The study introduces a new CRU system that combines speech recognition with surveillance info, enhancing call-sign detection in noisy environments.

Findings

Call-sign accuracy improved up to four times with surveillance data.

Data augmentation enhances performance on high WER transcripts.

Model adapts effectively to unseen airspaces.

Abstract

Air traffic control (ATC) relies on communication via speech between pilot and air-traffic controller (ATCO). The call-sign, as unique identifier for each flight, is used to address a specific pilot by the ATCO. Extracting the call-sign from the communication is a challenge because of the noisy ATC voice channel and the additional noise introduced by the receiver. A low signal-to-noise ratio (SNR) in the speech leads to high word error rate (WER) transcripts. We propose a new call-sign recognition and understanding (CRU) system that addresses this issue. The recognizer is trained to identify call-signs in noisy ATC transcripts and convert them into the standard International Civil Aviation Organization (ICAO) format. By incorporating surveillance information, we can multiply the call-sign accuracy (CSA) up to a factor of four. The introduced data augmentation adds additional performance on high WER transcripts and allows the adaptation of the model to unseen airspaces.