Lip-reading the BKB sentence lists: corrections for list and practice effects

Area of Science:

• Audiology
• Speech Perception
• Human Communication

Background:

• Assessing speech perception is crucial for understanding auditory and visual communication.
• Standardized sentence lists are vital tools for evaluating speech recognition abilities.
• Lip-reading accuracy can be influenced by various sensory and cognitive factors.

Purpose of the Study:

• To evaluate the impact of synchronized acoustic pulses on lip-reading performance.
• To quantify improvements in speech perception with practice and list exposure.
• To develop models for compensating practice effects and inter-list difficulty variations.

Main Methods:

• Two groups of 21 adults with normal hearing lip-read standard sentence lists.
• One group lip-read without sound (LR:alone), the other lip-read with synchronized acoustic pulses (LR&Lx).
• Performance was measured using loose (KW(L)) and tight (KW(T)) keyword scoring, analyzed with logistic regression.

Main Results:

• Lip-reading with acoustic pulses (LR&Lx) yielded significantly better performance than lip-reading alone (LR:alone).
• Speech perception scores improved linearly with the logarithm of list presentation order in both conditions.
• Higher initial scores correlated with greater improvement through experience with the lists.

Conclusions:

• Synchronized acoustic pulses enhance lip-reading effectiveness in speech perception.
• Practice effects and inter-list difficulty can be modeled and compensated for, improving assessment accuracy.
• The findings provide a framework for more precise evaluation of speech recognition abilities.