The ISCSLP 2022 Intelligent Cockpit Speech Recognition Challenge (ICSRC): Dataset, Tracks, Baseline and Results

TL;DR

The paper presents the ISCSLP 2022 ICSRC, a challenge focused on speech recognition in intelligent vehicle cockpits, introducing a new dataset, track setups, baseline systems, and analyzing system performances.

Contribution

It introduces a new cockpit speech dataset, defines two challenge tracks for resource-constrained and unconstrained systems, and provides baseline systems and analysis of submitted results.

Findings

Resource constraints significantly affect system performance.

Baseline systems establish reference points for future research.

Diverse acoustic conditions challenge speech recognition robustness.

Abstract

This paper summarizes the outcomes from the ISCSLP 2022 Intelligent Cockpit Speech Recognition Challenge (ICSRC). We first address the necessity of the challenge and then introduce the associated dataset collected from a new-energy vehicle (NEV) covering a variety of cockpit acoustic conditions and linguistic contents. We then describe the track arrangement and the baseline system. Specifically, we set up two tracks in terms of allowed model/system size to investigate resource-constrained and -unconstrained setups, targeting to vehicle embedded as well as cloud ASR systems respectively. Finally we summarize the challenge results and provide the major observations from the submitted systems.