End-Point Detection with State Transition Model based on Chunk-Wise Classification

TL;DR

This paper introduces a robust end-point detection method using a chunk-wise classification-based state transition model that reduces errors caused by noisy environments, improving speech/non-speech detection accuracy.

Contribution

It proposes a novel chunk-wise classification approach for state transition modeling in end-point detection, enhancing robustness against VAD errors in noisy conditions.

Findings

Improved accuracy in noisy environments.

Reduced false transitions due to chunk-wise aggregation.

Lower phone error rate in evaluations.

Abstract

A state transition model (STM) based on chunk-wise classification was proposed for end-point detection (EPD). In general, EPD is developed using frame-wise voice activity detection (VAD) with additional STM, in which the state transition is conducted based on VAD's frame-level decision (speech or non-speech). However, VAD errors frequently occur in noisy environments, even though we use state-of-the-art deep neural network based VAD, which causes the undesired state transition of STM. In this work, to build robust STM, a state transition is conducted based on chunk-wise classification as EPD does not need to be conducted in frame-level. The chunk consists of multiple frames and the classification of chunk between speech and non-speech is done by aggregating the decisions of VAD for multiple frames, so that some undesired VAD errors in a chunk can be smoothed by other correct VAD decisions. Finally, the model was evaluated in both qualitative and quantitative measures including phone error rate.