Unsupervised Uncertainty Measures of Automatic Speech Recognition for Non-intrusive Speech Intelligibility Prediction

TL;DR

This paper introduces an unsupervised method leveraging ASR uncertainty estimates to predict speech intelligibility without needing reference signals or labels, outperforming traditional intrusive methods.

Contribution

It presents a novel unsupervised approach using ASR uncertainty for speech intelligibility prediction, eliminating the need for labeled data or clean references.

Findings

Uncertainty from ASR models correlates strongly with speech intelligibility.

The method outperforms traditional intrusive intelligibility prediction methods.

Validated on two speech databases with promising results.

Abstract

Non-intrusive intelligibility prediction is important for its application in realistic scenarios, where a clean reference signal is difficult to access. The construction of many non-intrusive predictors require either ground truth intelligibility labels or clean reference signals for supervised learning. In this work, we leverage an unsupervised uncertainty estimation method for predicting speech intelligibility, which does not require intelligibility labels or reference signals to train the predictor. Our experiments demonstrate that the uncertainty from state-of-the-art end-to-end automatic speech recognition (ASR) models is highly correlated with speech intelligibility. The proposed method is evaluated on two databases and the results show that the unsupervised uncertainty measures of ASR models are more correlated with speech intelligibility from listening results than the predictions made by widely used intrusive methods.