Abstract

Spoken word recognition is affected by many factors including sound pressure level, signal-to-noise ratio, word familiarity, word frequency, and phoneme sequence plausibility. Of these factors, word familiarity is assumed to have a strong effect but has received little attention. One reason for this is that word familiarity data for most words have been unavailable. However, we can now confirm the word familiarity effect because in 1999 Amano and Kondo published a word familiarity database for about 80<th>000 Japanese words. Experimental results clearly show that spoken words with a high familiarity are recognized better and faster than those with a low familiarity. In addition, word familiarity correlates with the recognition score of spoken words much better than word frequency, which was previously regarded as an influential factor in word recognition. Because the word familiarity effect is so strong, it is very important to control the familiarity of spoken words to obtain stable and reasonable results when assessing the speech hearing ability of individuals, and when measuring speech clarity in communication media and public spaces.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call