What Is a Word Recognition Test?

Article At a Glance

• Word recognition testing measures how accurately you repeat spoken words under controlled test conditions. It provides information that a pure-tone audiogram alone cannot provide.
• A Word Recognition Score (WRS) is usually reported as the percentage of test words repeated correctly.
• Speech Reception Threshold (SRT) and word recognition testing measure different aspects of hearing. SRT estimates the level at which speech can be recognized, while word recognition testing examines accuracy when speech is presented at an audible level.
• WRS results depend on the test materials, presentation level, hearing-loss pattern, test method, and individual patient. Scores should therefore be interpreted in clinical context.
• A WRS can help guide hearing-care decisions, but it does not by itself determine hearing-aid benefit or cochlear-implant candidacy.

A standard hearing test tells you a great deal about the softest sounds you can detect across different frequencies.

But hearing thresholds are only one part of hearing.

You can detect that someone is speaking and still have difficulty understanding exactly what they said.

That is one reason audiologists use speech testing as part of many comprehensive hearing evaluations.

A word recognition test examines how accurately a person can repeat spoken words under controlled listening conditions.

The resulting Word Recognition Score provides information about speech understanding that complements the pure-tone audiogram.

What Is a Word Recognition Test?

During word recognition testing, a patient listens to a standardized set of words and repeats each word they hear.

The clinician records how many responses are correct.

The result is generally reported as a percentage.

For example, if 25 words are presented and the patient correctly repeats 20, the score is:

20 ÷ 25 × 100 = 80%

That percentage describes performance for that particular test condition.

It should not automatically be interpreted as meaning the person understands 80% of everything said in everyday life.

Real-world communication can be much more complicated because it may involve:

• Background noise
• Multiple speakers
• Distance
• Room reverberation
• Rapid speech
• Unfamiliar accents
• Reduced visual cues
• Attention and fatigue

Word recognition testing is therefore one useful clinical measurement rather than a complete simulation of everyday conversation.

How Word Recognition Testing Works

Word recognition testing is commonly performed using recorded speech materials or monitored live voice, depending on the clinic, test protocol, patient, and clinical question.

Speech may be delivered through:

• Insert earphones
• Headphones
• Loudspeakers in appropriate test situations

The words are presented at a level selected by the clinician to evaluate speech recognition appropriately for that patient.

There is no single presentation level that is correct for every person.

The appropriate level may depend on factors such as:

• Hearing thresholds
• Speech Reception Threshold
• Comfortable listening levels
• Degree and configuration of hearing loss
• Previous speech-test results
• The diagnostic question being investigated

The clinician then compares the patient’s responses with the test words and calculates the percentage recognized correctly.

Are Word Lists Always 10 to 25 Words Long?

No.

Clinical speech-recognition protocols use different standardized word lists and list lengths.

Some testing may use shortened lists when appropriate, while other situations call for longer lists to improve confidence in the result.

The number of words matters because a percentage based on a small sample can be less precise than one based on a larger sample.

For example, on a 10-word list, one incorrect response changes the score by 10 percentage points.

On a 50-word list, one incorrect response changes the score by only 2 percentage points.

Clinicians balance test reliability with factors such as patient age, attention, fatigue, available test materials, and the purpose of the evaluation.

There is therefore no universal 10-to-25-word “sweet spot” that applies to every word recognition test.

What Does a Word Recognition Score Mean?

A WRS describes how accurately a patient recognized the test words under the conditions in which they were presented.

In general, higher scores indicate better word recognition under those conditions, while lower scores indicate greater difficulty recognizing the speech material.

However, a WRS should not be interpreted using one universal set of percentage categories.

Different clinics, researchers, test materials, and clinical protocols may classify results differently.

The interpretation also depends on:

• The patient’s pure-tone hearing thresholds
• The presentation level used
• Which ear was tested
• The word list used
• Whether recorded or live-voice presentation was used
• Whether the score is consistent with previous testing
• Whether there is a meaningful difference between the ears
• The patient’s broader medical and hearing history

For this reason, a score such as 80% should not automatically be labeled “mild difficulty,” and a score below a particular percentage should not automatically be interpreted as proof that hearing aids will or will not help.

Why the Audiogram and WRS Can Tell Different Stories

Pure-tone audiometry and word recognition testing measure different things.

The audiogram primarily shows the softest test signals a person can detect across selected frequencies.

Word recognition testing examines the ability to identify speech material when it is presented at an appropriate audible level.

Two people can therefore have similar pure-tone thresholds but different word recognition results.

Possible reasons include differences in:

• The underlying auditory disorder
• Degree and configuration of hearing loss
• Speech presentation level
• Test reliability
• Age and health factors
• Language familiarity
• Attention
• Other auditory or neurological factors

A difference between pure-tone thresholds and speech-recognition performance can provide useful clinical information, but it does not by itself identify the exact cause.

Learn more about why you may hear sounds but have difficulty understanding words.

What Is Speech Reception Threshold?

The Speech Reception Threshold (SRT) is another common speech-audiometry measurement.

SRT generally estimates the lowest hearing level at which a person can correctly recognize approximately half of the speech material used for threshold testing.

In English-speaking adult testing, this commonly involves familiar two-syllable words with approximately equal stress on both syllables, known as spondees.

Examples can include words such as “baseball” or “hotdog.”

SRT can be compared with pure-tone thresholds as part of evaluating the consistency of the hearing-test results.

It does not measure the same thing as word recognition testing.

SRT vs. WRS

The distinction is important.

SRT asks approximately how soft speech can become before recognition reaches a threshold criterion.

WRS asks how accurately the patient recognizes a set of words at a selected audible presentation level.

These measurements complement one another.

Neither should be treated as a complete description of everyday communication ability.

Why Presentation Level Matters

Word recognition performance can change depending on how loudly the speech material is presented.

If words are presented too softly, performance may be reduced because parts of the speech signal are not sufficiently audible.

Increasing the presentation level may improve recognition for some listeners.

However, simply making speech louder does not always continue improving understanding.

Some hearing disorders can produce limited word recognition even when the speech is clearly audible.

For this reason, clinicians select presentation levels according to the patient’s hearing and the purpose of the test rather than using one fixed rule for everyone.

In some cases, word recognition may be measured at more than one presentation level.

Testing across several levels can provide additional information about how speech-recognition performance changes with intensity.

What Can Affect a Word Recognition Score?

Many factors can influence WRS results.

Degree and Configuration of Hearing Loss

The frequencies affected and the amount of hearing loss can influence how much speech information is available to the listener.

Type and Cause of Hearing Loss

Conductive, cochlear, neural, and other auditory disorders can affect speech recognition differently.

However, a low WRS does not automatically prove that the auditory nerve is damaged.

Additional clinical information may be necessary when neural or retrocochlear pathology is suspected.

Presentation Level

Speech that is not sufficiently audible may produce a lower score.

The clinician therefore considers hearing thresholds and other information when selecting the test level.

Test Materials

Different standardized word lists are not necessarily interchangeable.

The language, difficulty, and construction of the test materials can affect performance.

Recorded vs. Live-Voice Testing

Recorded materials provide more consistent presentation across patients and test sessions.

Monitored live voice may still be used in clinical practice in some situations, but speaker differences can introduce additional variability.

Language and Familiarity

A speech-recognition test depends partly on familiarity with the language and words being presented.

Results should be interpreted carefully when the test language is not the patient’s primary or fluent language.

Attention and Fatigue

Attention, understanding of the task, and fatigue can influence performance.

Clinicians consider these factors when determining whether a result appears reliable.

Does a Low WRS Mean Auditory-Nerve Damage?

Not necessarily.

A disproportionately poor speech-recognition score can sometimes raise questions about whether something beyond ordinary cochlear hearing loss should be investigated.

But WRS alone cannot diagnose auditory-nerve damage.

When findings are unusual, strongly asymmetrical, or inconsistent with the rest of the hearing evaluation, the audiologist or medical clinician may consider whether additional assessment is appropriate.

The decision depends on the complete clinical picture.

Can Someone Have Normal Pure-Tone Thresholds and Poor Speech Recognition?

Yes, although the interpretation requires care.

Pure-tone thresholds measure sensitivity to relatively simple test signals.

Speech understanding is more complex.

Some auditory disorders can interfere with speech perception even when conventional pure-tone thresholds are relatively good.

Examples can include certain neural auditory disorders and other conditions affecting auditory processing.

However, one unexpectedly low WRS should not automatically be labeled auditory neuropathy or a central auditory processing disorder.

The result should first be evaluated for:

• Test reliability
• Presentation level
• Language factors
• Word-list effects
• Asymmetry
• Other audiological findings
• Relevant medical history

Additional testing may be appropriate depending on the findings.

Does WRS Predict Whether Hearing Aids Will Work?

WRS can help clinicians discuss expectations, but it does not directly determine whether hearing aids will be beneficial.

A person with relatively strong word recognition may have good potential to use amplified speech information when it is made appropriately audible.

A person with poorer word recognition may experience less clarity from amplification alone.

But real-world hearing-aid benefit depends on many additional factors, including:

• Degree and configuration of hearing loss
• Speech understanding in quiet and noise
• Hearing-aid fitting and verification
• Listening environment
• Communication needs
• Use of directional microphones or remote microphones
• Visual cues
• Experience with amplification
• Individual preferences and goals

A low unaided WRS therefore does not mean hearing aids are useless, while a high WRS does not guarantee excellent hearing-aid performance in every environment.

Does a Hearing Aid Improve the Word Recognition Score?

This question requires an important distinction.

A diagnostic WRS is commonly measured under controlled conditions without hearing aids.

A hearing aid can improve audibility by making speech information more accessible.

That can improve communication when lack of audibility is an important part of the problem.

But amplification does not necessarily change the auditory system’s underlying maximum ability to distinguish speech.

Hearing-aid benefit is therefore better evaluated using appropriate aided testing, real-ear verification, speech-in-noise measures when indicated, and reports of real-world communication rather than assuming that the unaided diagnostic WRS itself should increase.

What About Hearing Aids in Both Ears?

When both ears have aidable hearing loss, bilateral amplification may provide advantages for some people, including access to sound from both sides and support for binaural listening.

However, bilateral hearing aids should not be promised to increase a particular WRS percentage.

The appropriate fitting strategy depends on the individual’s hearing loss, speech-recognition results, communication needs, medical findings, and preferences.

Can WRS Determine Cochlear Implant Candidacy?

No single unaided WRS percentage determines cochlear-implant candidacy.

Cochlear-implant evaluation uses specific clinical criteria and commonly considers speech-recognition performance with appropriately fitted hearing aids, along with the degree of hearing loss and other factors.

Candidacy criteria can also differ according to:

• Age
• Device and regulatory indications
• Insurance or healthcare-system requirements
• Whether one or both ears are affected
• The specific speech-recognition materials used
• Performance with appropriately fitted amplification

Someone with poor word recognition should not assume that they automatically need a cochlear implant.

Likewise, people should not assume they are ineligible based on one percentage they see on an audiogram report.

When hearing aids provide limited speech understanding despite appropriate fitting, a hearing professional can determine whether a cochlear-implant evaluation or another rehabilitation option is worth considering.

Why Speech-in-Noise Testing May Also Matter

Traditional word recognition testing is often performed in quiet.

That can be very different from everyday listening.

Many people report that they communicate reasonably well in quiet but struggle substantially in:

• Restaurants
• Family gatherings
• Meetings
• Vehicles
• Busy workplaces
• Other environments with competing sound

Speech-in-noise testing can provide additional information about these difficulties.

A person can have a relatively strong WRS in quiet and still have substantial difficulty understanding speech in noise.

Learn more about speech-in-noise testing.

Why Word Recognition Scores Should Be Compared Carefully Over Time

A change in WRS between hearing tests may be clinically meaningful, but percentages should not automatically be compared as though every small difference represents a true change in hearing.

Word recognition testing involves statistical variability.

The number of words presented, test list, presentation level, test method, and previous score all affect how confidently two results can be considered different.

For example, a change from 88% to 84% does not necessarily mean speech understanding has deteriorated.

Clinicians can determine whether the difference is large enough to be meaningful in the context of the test protocol.

Frequently Asked Questions

What Is a Word Recognition Test Used For?

A word recognition test measures how accurately a person identifies spoken words under controlled test conditions.

It complements pure-tone audiometry by providing information about speech recognition rather than sound detection alone.

The result can help audiologists characterize hearing ability, compare the ears, evaluate consistency with other findings, discuss rehabilitation expectations, and determine whether additional assessment may be appropriate.

What Is a Good Word Recognition Score?

In general, higher scores indicate better recognition of the particular speech material under the conditions used.

However, there is no single universal percentage table that should be used by patients to classify every WRS as normal, mild, moderate, poor, or very poor.

Interpretation depends on the test materials, presentation level, hearing thresholds, age and language factors, differences between the ears, and other clinical findings.

Your audiologist can explain what the score means in the context of your complete hearing evaluation.

How Is a Word Recognition Score Calculated?

The basic calculation is:

Number of correctly repeated words ÷ total number of words presented × 100.

For example:

• 20 correct responses out of 25 words = 80%
• 40 correct responses out of 50 words = 80%

Although both examples produce the same percentage, the statistical precision of the measurements is not necessarily identical because the list lengths differ.

That is one reason clinicians interpret the percentage together with the test protocol.

How Many Words Are Used in a Word Recognition Test?

There is no single number used for every patient.

Standardized speech materials may contain different list lengths, and clinicians may use full or shortened lists depending on the clinical situation.

Longer lists can provide greater confidence in the score, while shorter testing may sometimes be appropriate for practical or clinical reasons.

Can You Have Relatively Good Hearing Thresholds and Still Have a Low WRS?

Yes.

Pure-tone thresholds and speech recognition measure different aspects of auditory function.

However, an unexpectedly poor WRS should not automatically be attributed to auditory-nerve damage.

The clinician considers the complete test pattern, reliability, presentation level, language factors, asymmetry, and other findings before determining whether additional evaluation is warranted.

What Is the Difference Between SRT and WRS?

The Speech Reception Threshold estimates the level at which a person recognizes approximately half of the speech material used for threshold testing.

The Word Recognition Score measures the percentage of test words correctly identified at a selected audible presentation level.

SRT focuses on a speech-recognition threshold.

WRS focuses on recognition accuracy under the selected test condition.

Does a Low Word Recognition Score Mean Hearing Aids Will Not Help?

No.

A low WRS can indicate that amplification alone may not provide perfect speech clarity, but it does not prove that hearing aids will provide no benefit.

Hearing aids may still improve audibility, environmental awareness, communication in some situations, and access to assistive technology.

The amount of benefit should be assessed using the person’s complete hearing profile and real-world communication needs.

Does a High WRS Guarantee Hearing Aids Will Work Well?

No.

A high WRS under controlled quiet conditions is encouraging information, but everyday listening can involve background noise, distance, reverberation, competing speakers, and other challenges.

Hearing-aid benefit also depends on appropriate fitting, verification, listening environments, communication needs, and individual factors.

Does Poor Word Recognition Mean I Need a Cochlear Implant?

Not automatically.

Poor speech recognition can be one reason a clinician considers whether additional rehabilitation options should be evaluated.

Cochlear-implant candidacy, however, involves more than an unaided WRS and commonly includes aided speech-recognition testing and other clinical criteria.

A cochlear-implant center can perform a formal candidacy evaluation when appropriate.

The Bottom Line

A Word Recognition Score adds important information to a hearing evaluation because detecting sound and understanding speech are not the same thing.

The audiogram measures hearing sensitivity across frequencies.

Word recognition testing measures how accurately you identify speech material under controlled test conditions.

SRT, WRS, pure-tone thresholds, speech-in-noise performance, medical history, and real-world communication difficulties can all contribute different pieces of information.

The most important point is not to treat one WRS percentage as a diagnosis or a prediction of exactly how well a hearing aid or cochlear implant will work.

A qualified hearing professional can interpret the score in the context of the complete evaluation and use it to help determine which additional testing, hearing technology, communication strategies, or medical follow-up may be appropriate.