Word Recognition Score (WRS)

My recent tonal test took about 4 minutes and the WRS test took about one minute at my ENT’s office audioloigist. Total exam.

The two previous WRS tests, there, were not recorded input – but done live. My previous two tests showed WRS with problems and were done live by the same person.-- a woman. The third and most recent test showed no problems at all with WRS (100%) but also done live by a different female audiologist. (My hearing loss is in the upper frequencies)…

My personal experience outside that office shows without a doubt I am having problems understanding certain words. Can two different audiologists with different voice frequencies generate different word recognition scores? Should they have used a recorded test input for comparison purposes?.

Recordings are consistent.
Your tests were not.


My initial audiogram and testing used recorded sounds and generated tones. It took probably 20 minutes or more.

1 Like

I had a test on Tuesday, tones through headphones then bone conduction and that took a good 20 minutes and wasn’t particularly drawn out. I can’t see word recognition scores being consistent with different voices even if they were both female, were they done at the same office? Doesn’t seem very reproducible to me but would also be dependant on your audiogram

WRS is variable even when taped, and different audiologists are often testing for different things when doing it and present it at different levels. Live-voice WRS is only going to give a very gross measure, and becomes more and more meaningless as it is done by different speakers. But the ENT probably isn’t looking for how well you are able to hear speech in quiet. She’s looking for whether you have 100% in one ear and 0% in the other which could indicate some sort of retrocochlear lesion.

Was she looking at you when the test was done?

People don’t typically realize how much they read lips, and with high frequency loss, the sound you miss are formed by the teeth, tongue, and lips, and easy to recognize. When I do live voice discrim, I always cover my mouth so that lip reading is not possible.

1 Like

@eric.cobb I can remember going for WRS and the grad student turned her back to me. :flushed: With my profound high frequency loss I badly failed that particular WRS. So badly I asked her to do it again but to get a piece of paper and cover her mouth… My scores still weren’t the best but they were a lot better than 2% both ears.

Can someone help me to read parts of my audiogram? I got the “hearing level” thing, but can’t figure out the other numbers… there is a box labeled: "PTA (dBHL) /AI (%), with readings for Right and left … with categories AC, BC and IA … for Right it shows AC 45; AI 14; Left: AC 55, AI 9. is this word recognition? I really don’t see anything in WRS/SRS box, except for 60 db Hl for both ears? Does this mean I didn’t have a word recognition test?

On an older audiogram, Under “speech audiometry” box, it showed the following but again, I don’t really know how to read it…

SRT/SAT Word Recog

R 45 dB 70/72 %

L 50 dB 80/56 %

I’m confused at to the two percentages? Or is that one percentage?

THanks for any help… I know I should go back to audiologist for more thorough explanation but am trying to get as smart as I can before I go…

The SRT is Speech Reception Threshold and is the lowest level at which you get 50% of the spondiac (two syllable words) correct. The 70/72% is a 72% word recognition score when presented at 70dB. The right and left are a little different which is very common. The AC PTA is air conduction (with headphones or insert earphones) pure tone average of 500, 1000 and 2000 hertz. The AI is articulation index and represents the number of data points out of 100 that fall within you range of hearing. The data points are weighted with more of them in the critical high frequency area. By the way, WRS is also commonly called PB Max (phonetically balanced words maximum score} or also word discrimination score. There can be big differences in scores with regards to different voices and male or female presenters. If they have not done it yet, they should do speech mapping and hearing in noise tests as well.