Study on the Effect of Face Masks on Forensic Speaker Recognition

The COVID-19 pandemic has led to a dramatic increase in the use of face masks. Face masks can affect both the acoustic properties of the signal and the speech patterns and have undesirable effects on automatic speech recognition systems as well as on forensic speaker recognition and identification systems. This is because the masks introduce both intrinsic and extrinsic variability into the audio signals. Moreover, their filtering effect varies depending on the type of mask used. In this paper we explore the impact of the use of different masks on the performance of an automatic speaker recognition system based on Mel Frequency Cepstral Coefficients to characterise the voices and on Support Vector Machines to perform the classification task. The results show that masks slightly affect the classification results. The effects vary depending on the type of mask used, but not as expected, as the results with FPP2 masks are better than those with surgical masks. An increase in speech intensity has been found with the FPP2 mask, which is related to the increased vocal effort made to counteract the effects of hearing loss. © 2022, Springer Nature Switzerland AG.

Automatic Rephrasing of Transcripts-based Action Items

The automated transcription of spoken language, and meetings, in particular, is becoming more widespread as automatic speech recognition systems are becoming more accurate. This trend has significantly accelerated since the outbreak of the COVID-19 pandemic, which led to a major increase in the number of online meetings. However, the transcription of spoken language has not received much attention from the NLP community compared to documents and other forms of written language. In this paper, we study a variation of the summarization problem over the transcription of spoken language: given a transcribed meeting, and an action item (i.e., a commitment or request to perform a task), our goal is to generate a coherent and self-contained rephrasing of the action item. To this end, we compiled a novel dataset of annotated meeting transcripts, including human rephrasing of action items. We use state-of-the-art supervised text generation techniques and establish a strong baseline based on BART and UniLM (two pretrained transformer models). Due to the nature of natural speech, language is often broken and incomplete and the task is shown to be harder than an analogous task over email data. Particularly, we show that the baseline models can be greatly improved once models are provided with additional information. We compare two approaches: one incorporating features extracted by coreference-resolution. Additional annotations are used to train an auxiliary model to detect the relevant context in the text. Based on the systematic human evaluation, our best models exhibit near-human-level rephrasing capability on a constrained subset of the problem. © 2021 Association for Computational Linguistics