# DNN adaptation by automatic quality estimation of ASR hypotheses

## TL;DR

This paper introduces a method for unsupervised DNN adaptation in ASR by using automatic quality estimation to select high-quality data, leading to significant performance improvements over baseline models.

## Contribution

It presents a novel approach that leverages automatic quality estimation to perform data selection for DNN adaptation in ASR, without requiring manual transcriptions.

## Key findings

- QE predictions closely match oracle adaptation results
- QE-driven adaptation significantly outperforms baseline models
- Method effective in both oracle and realistic conditions

## Abstract

In this paper we propose to exploit the automatic Quality Estimation (QE) of ASR hypotheses to perform the unsupervised adaptation of a deep neural network modeling acoustic probabilities. Our hypothesis is that significant improvements can be achieved by: i)automatically transcribing the evaluation data we are currently trying to recognise, and ii) selecting from it a subset of "good quality" instances based on the word error rate (WER) scores predicted by a QE component. To validate this hypothesis, we run several experiments on the evaluation data sets released for the CHiME-3 challenge. First, we operate in oracle conditions in which manual transcriptions of the evaluation data are available, thus allowing us to compute the "true" sentence WER. In this scenario, we perform the adaptation with variable amounts of data, which are characterised by different levels of quality. Then, we move to realistic conditions in which the manual transcriptions of the evaluation data are not available. In this case, the adaptation is performed on data selected according to the WER scores "predicted" by a QE component. Our results indicate that: i) QE predictions allow us to closely approximate the adaptation results obtained in oracle conditions, and ii) the overall ASR performance based on the proposed QE-driven adaptation method is significantly better than the strong, most recent, CHiME-3 baseline.