Qualitative Evaluation of Language Model Rescoring in Automatic Speech Recognition

Abstract

Evaluating automatic speech recognition (ASR) systems is a classical but difficult and still open problem, which often boils down to focusing only on the word error rate (WER). However, this metric suffers from many limitations and does not allow an in-depth analysis of automatic transcription errors. In this paper, we propose to study and understand the impact of rescoring using language models in ASR systems by means of several metrics often used in other natural language processing (NLP) tasks in addition to the WER. In particular, we introduce two measures related to morpho-syntactic and semantic aspects of transcribed words: 1) the POSER (Part-of-speech Error Rate), which should highlight the grammatical aspects, and 2) the Em-bER (Embedding Error Rate), a measurement that modifies the WER by providing a weighting according to the semantic distance of the wrongly transcribed words. These metrics illustrate the linguistic contributions of the language models that are applied during a posterior rescoring step on transcription hypotheses.

HATS: An Open Data Set Integrating Human Perception Applied to the Evaluation of Automatic Speech Recognition Metrics

“…Next, we examine three semantic metrics based on word embedding representations. The first one, Embedding Error Rate (EmbER) [2], is a WER where substitution errors are weighted according to the cosine distance between the reference and the substitute word embeddings obtained from fastText [14,3]. The second one, SemDist [20], involves calculating the cosine similarity between the reference and hypothesis using embeddings obtained at the sentence level.…”

“…Consequently, there has been a growing interest in developing new metrics to evaluate ASR systems. Some researchers [23,27,13,20,2] have therefore started exploring alternative metrics that can more accurately assess the quality and effectiveness of automatic transcriptions. Similarly, these issues have been observed in the field of machine translation.…”