Comparison of Acceptable Noise Level Generated Using Different Transducers and Response Modes

Abstract

The acceptable noise level (ANL) was defined by subtracting the background noise level (BNL) from the most comfortable listening level (MCL) (ANL = MCL − BNL). This study compared the ANL obtained through different methods in 20 Chinese subjects with normal hearing. ANL was tested with Mandarin speech materials using a loudspeaker or earphones, with each subject tested by himself or by the audiologist. The presentation and response modes were as follows: (1) loudspeaker with self-adjusted noise levels using audiometer controls (LS method); (2) loudspeaker with the subject signaling the audiologist to adjust speech and noise levels (LA method); (3) earphones with self-adjusted noise levels using audiometer controls (ES method); and (4) earphones with the subject signaling the audiologist to adjust speech and noise levels (EA method). ANL was calculated from three measurements with each method. There was no significant difference in the ANL obtained through different presentation modes or response modes sound. The correlations between ANL, MCL, and BNL obtained from each two methods were significant. In conclusion, the ANL in normal-hearing Mandarin listeners may not be affected by presentation modes such as a loudspeaker or earphones nor is it affected by self-adjusted or audiologist-adjusted response modes. Earphone audiometry is as reliable as sound field audiometry and provides an easy and convenient way to measure ANL.

The Impact of Transducer Selection on the Acceptable Noise Level

“…The ANL literature has one report of transducer comparisons from Xia et al 25 As in the present study, they reported no difference in ANL for circumaural earphones versus loudspeaker. Similarly, in their study they identified MCL differences by transducer with circumaural earphones producing a higher mean MCL (mean across participants for trial 1 was 44.8 dB, trial 2 was 45.2 dB, and trial 3 was 43.8 dB) than when evaluated with the loudspeaker (mean across participants for trial 1 was 38.1 dB, trial 2 was 38.1 dB, and trial 3 was 38.6 dB).…”

“…While it is difficult to identify phrases in the twelve-talker babble, individual words can be identified. It is possible that the low ANL findings for the Xia et al 25 study were due to the background stimulus not producing adequate interference. Implementation of speech stimuli of two languages would likely cause the two auditory streams to be segregated by the listener more readily, producing very low ANLs.…”

“…It is likely that the listeners used the speech at MCL as an anchor point and that BNL was set relative to the speech, producing consistent ANLs across transducers. t should also be noted that Xia et al 25 documented ANLs that were much lower than those typically identified. In their study, ANLs ranged from À 1.1 to 0.6 dB rather than falling closer to 10 or 11 dB.…”

“…It should also be noted that Xia et al 25 documented ANLs that were much lower than those typically identified. In their study, ANLs ranged from À1.1 to 0.6 dB rather than falling closer to 10 or 11 dB.…”

Acceptable Noise Level Stability Over a One-Year Period of Time

“…ANLs recorded during a session are reliable and repeatable (Nabelek et al; 23 Freyaldenhoven et al; 8 Gordon-Hickey et al; 11 Xia et al). 30 Furthermore, the measurements of MCLs and background noise levels (BNLs) have been found to be reliable (Nabelek et al; 23 Freyaldenhoven and Smiley; 8 Freyaldenhoven et al; 8 Mueller et al; 20 Nabelek et al). 20,21 Interrater reliability of ANLs is strong (Gordon-Hickey et al).…”