On-Device Constrained Self-Supervised Speech Representation Learning for Keyword Spotting via Knowledge Distillation

TL;DR

This paper introduces a knowledge distillation approach for self-supervised speech representation learning tailored for on-device keyword spotting, achieving high accuracy under resource constraints and noisy conditions.

Contribution

It presents a novel teacher-student framework with dual-view cross-correlation distillation for efficient on-device keyword spotting.

Findings

High accuracy in keyword detection on large in-house dataset

Robust performance in noisy environments

Effective model compression for on-device deployment

Abstract

Large self-supervised models are effective feature extractors, but their application is challenging under on-device budget constraints and biased dataset collection, especially in keyword spotting. To address this, we proposed a knowledge distillation-based self-supervised speech representation learning (S3RL) architecture for on-device keyword spotting. Our approach used a teacher-student framework to transfer knowledge from a larger, more complex model to a smaller, light-weight model using dual-view cross-correlation distillation and the teacher's codebook as learning objectives. We evaluated our model's performance on an Alexa keyword spotting detection task using a 16.6k-hour in-house dataset. Our technique showed exceptional performance in normal and noisy conditions, demonstrating the efficacy of knowledge distillation methods in constructing self-supervised models for keyword spotting tasks while working within on-device resource constraints.