Making the Case for Testing Soft Speech and Speech in Noise

Jane Madell
May 28, 2019

Everyone with hearing loss complains about the ability to hear and understand in the presence of competing noise. Unfortunately, audiologists do not test in noise often, and pediatric audiologists test even less often.


Why doesn’t speech in  noise get tested?


Audiologists do the work we do to help people and we really do not want to upset them more than necessary, so we often don’t test in noise because we know that our patients will struggle and we don’t want to distress them. But, if we do not test in difficult conditions we rally cannot know when we need to change technology settings and help our patients develop realistic expectations.

If we want to know how a person is doing we absolutely need to test our patients ability to hear in difficult listening situations including soft speech (35 dBHL) and in competing noise.


Hearing soft speech


Children learn 90% of what they learn through incidental learning – by overhearing. If a child cannot easily hear soft speech how can we expect them to overhear? If a child has aided thresholds at 30 or 35 dB she will hear soft speech at a very soft level – like a whisper. This means she will miss most of what is said. Not good for learning!!

The goal is that soft speech be at a comfortably loud level. If thresholds are at 20-25 dB we can expect the child to be able to hear soft speech but if it is at a higher level the child will not hear soft speech.


Hearing in noise


Everyone agrees that the ability to hear in noise is critical. But we don’t all test it. Why? “It takes too much time” is not a good excuse. We need to know how our patients hear in noise.

Although the American Academy of Audiology recommends speech in noise testing, Beck and Benitz (2019) report that less than 15% of audiologists test in noise. We may not be able to do speech in noise testing for a 3 month old but once kids talk, and certainly with adults, we are able to do this testing. People with hearing loss need speech to be at least 8 dB louder than the background noise in order to repeat back 50% of the words. The noise should be 4 talker babble rather that speech noise, white noise, pink noise etc. They are not good representations of typical listening conditions.


Advice for families


You are your own best advocates. If your audiologist is not testing with the technology (hearing aids, CI’s, Baha’s, RM) ask her to do the testing. Ask for the following

  • Aided threshold testing with each piece of technology alone and with all those used together (eg. hearing aids and FM)
  • Speech perception in sound field with technology at normal conversational levels (50 dBHL) – right alone, left alone, binaural, binaural with FM
  • Speech perception in sound field with technology at soft conversational levels (35 dBHL) –binaural, binaural with FM
  • Speech perception in sound field with technology at normal conversational levels in competing noise (50 dBHL + 5 SNR) or SIN test –binaural, binaural with FM


What do we do with the information?


Only by testing in multiple conditions do we really get a good picture of how a child is hearing. Having this information will help determine if she is hearing well enough, and if not, help determine if we need to modify technology settings, change technology, or develop therapy programs.

Only with data can we fix problems. Kids can be stars only if we do our job well.



  1. Can the author provide specifics about tests that are commercially available? This seems very relevant to adult fitting as well!

  2. Testing for speech understanding deficits is always good, provided we understand the neuronal development within the brain on a continuous basis. Is the child exposed to social noise ( like in Asian communities?).
    Biologically there is not much we can do when the first layer neurotransmitters get affected within the inner hair cells, causing reduced sensitivity of soft speech sounds. When this is detected, the parents of the child tested need counseling in controlling noise at home. What most audiologists don’t understand ids the reverberation factor at homes, which can be adding as much as 12 dbA to the existing levels. Most dwellings have a lot of congestion and not enough ventilation to bleed off the echoes from TV’s.. It is these echoes that form the basis of reduced sensitivity,to soft sounds and voices, leading to a sustained hearing loss in the first tier neurotransmitter zones within the hair cell receptors. Parents need to be told that their indulgences can destroy a child’s hearing significantly.
    The use of earbuds, iPods, can be lethal to hearing acuity, and must be discouraged, and again this is the parent’s responsibility to understand the consequences of connectivity, and high sound pressures within the ear canals. Testing is proof of the problem and greatly helps in reinforcement.The first indication of soft sound desensitization is an appearance of lack of interest in education (school), This is the best clue in an investigation. Use all tools available to subjectively understand the hearing situation, because psychology is the best tool to deal with hearing loss, and not hearing aids.!

  3. Please note that language heard as a learning curve event is not a sound, rather it’s speech coding. Coding is the essential part of speech and language development. Children have excellent inhibitive functions that allow GABA to precisely control the inputs of language in the presence of competing speech. Speech in noise testing allows this feature to be understood clinically and measured objectively, as also the results allow us to forecast the likely presence of any pathology that could reduce hearing acuity in the forthcoming years. The predictive element could help in future pharma applications and counseling strategies that would help understand the consequences of noise induced hearing loss, possibly right from its inception !

    Jay Muhury Hs.D
    Auditory Researcher.

Leave a Reply