Microscopic and Blind Prediction of Speech Intelligibility: Theory and Practice

Karbasi, Mahdie; Zeiler, Steffen; Kolossa, Dorothea

doi:10.1109/taslp.2022.3184888

Search citation statements

Order By: Relevance

Paper Sections

Select...

Problem Definition1

Citation Types

Supporting

Mentioning

Contrasting

Year Published

2022

2023

Publication Types

Select...

Article3

Relationship

Self Cite0

Independent3

Authors

Journals

Cited by 3 publications

(1 citation statement)

References 52 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Therefore, while we assume that the text corpus is known, the exact subset of the corpus used to evaluate intelligibility in each acoustic condition is unknown. In the present work, we only consider macroscopic SIP algorithms 15,16 that predict the average intelligibility of an acoustic condition and do not inherently model corpus predictability (such as SIP using automatic speech recognition 17–19 ).…”

Section: Problem Definitionmentioning

confidence: 99%

Modeling the effect of linguistic predictability on speech intelligibility prediction

Edraki

Chan

Fogerty

et al. 2023

JASA Express Letters

View full text Add to dashboard Cite

Many existing speech intelligibility prediction (SIP) algorithms can only account for acoustic factors affecting speech intelligibility and cannot predict intelligibility across corpora with different linguistic predictability. To address this, a linguistic component was added to five existing SIP algorithms by estimating linguistic corpus predictability using a pre-trained language model. The results showed improved SIP performance in terms of correlation and prediction error over a mixture of four datasets, each with a different English open-set corpus.

show abstract

Section: Problem Definitionmentioning

confidence: 99%