Auto-KWS 2021 Challenge: Task, Datasets, and Baselines

TL;DR

The Auto-KWS 2021 challenge promotes AutoML solutions for personalized, environment-robust keyword spotting, emphasizing real-world scenarios and speaker-specific keywords, with baseline systems provided for benchmarking.

Contribution

It introduces a new challenge focusing on automated, personalized keyword spotting in realistic environments, with datasets and baseline systems to foster development.

Findings

Challenge encourages development of personalized KWS systems.

Realistic environment datasets improve robustness.

Baseline systems serve as performance references.

Abstract

Auto-KWS 2021 challenge calls for automated machine learning (AutoML) solutions to automate the process of applying machine learning to a customized keyword spotting task. Compared with other keyword spotting tasks, Auto-KWS challenge has the following three characteristics: 1) The challenge focuses on the problem of customized keyword spotting, where the target device can only be awakened by an enrolled speaker with his specified keyword. The speaker can use any language and accent to define his keyword. 2) All dataset of the challenge is recorded in realistic environment. It is to simulate different user scenarios. 3) Auto-KWS is a "code competition", where participants need to submit AutoML solutions, then the platform automatically runs the enrollment and prediction steps with the submitted code.This challenge aims at promoting the development of a more personalized and flexible keyword spotting system. Two baseline systems are provided to all participants as references.