57 Speech Intelligibility Statistics

by

These speech intelligibility figures cover the measurement periods and populations identified in the available source material. Estimates, surveys and forecasts retain their reported scope.

Key Speech Intelligibility Statistics

The main findings are:

  • The Speech Transmission Index (STI) scale spans 0 to 1.0, where 0 represents a very poor channel and 1.0 an ideal channel with excellent speech transmission (2016 study/report context; IEC STI framework).
  • At STI 0–0.30, the reference relationship gives 0–34% syllable, 0–67% word and 0–89% sentence intelligibility (IEC 60268-16 relationship; reference STI relationship).
  • A Belo Horizonte study measured speech intelligibility in 18 classrooms across nine public schools (2012 measurements; Belo Horizonte, Brazil).
  • In a television study, 15 adults aged 59–82 with bilateral sensorineural hearing loss completed repeated speech-recognition tests (published study; older adults with hearing loss).
  • Nearly 2.5 billion people will have some degree of hearing loss by 2050 (2050 projection).
  • The Global Burden of Disease 2019 analysis estimated 1.57 billion people had hearing loss in 2019, with a 95% uncertainty interval of 1.51–1.64 billion.
  • STI 0–0.30 is classified as bad quality (IEC 60268-16 classification; IEC STI framework).
  • At STI 0.30–0.45, the reference relationship gives 34–48% syllable intelligibility (IEC 60268-16 relationship; reference STI relationship).
  • The classroom study tested 273 students, with a mean age of 9.4 years (2012 measurements; Belo Horizonte, Brazil).
  • Among young normal-hearing pilot listeners, unc captioned televised speech lists ranged from 86% to 98% correct, with a 92% mean (pilot testing; young normal-hearing listeners).
  • WHO projects that more than 700 million people will require hearing rehabilitation by 2050 (2050 projection).
  • The same GBD analysis estimated 43.45 million years lived with disability from hearing loss in 2019, with a 95% uncertainty interval of 29.68–61.80 million.
  • STI 0.30–0.45 is classified as poor quality (IEC 60268-16 classification; IEC STI framework).
  • At STI 0.30–0.45, the reference relationship gives 67–78% word intelligibility (IEC 60268-16 relationship; reference STI relationship).
  • Boys represented 45.4% of the tested students (2012 measurements; Belo Horizonte, Brazil).

Contents

Speech intelligibility statistics by STI quality band

STI is an objective channel metric, not a universal direct percentage for every listener or language.

Classification bands summarize a standards relationship rather than one experiment.

Classification band.

Non-human, unidirectional testing of selected products.

Secondary report of a prior study and dependent on test conditions.

Secondary report; average spans several devices and conditions.

Selected respirator and laboratory setup.

Suggested baseline, not a universally adopted performance requirement.

Speech intelligibility scores for syllables, words and sentences

Reference conversion varies with language, material and listener.

Reference conversion.

Speech intelligibility statistics in rooms, classrooms and hospitals

Cross-sectional study of selected public schools.

Study sample, not all Brazilian students.

Sample composition statistic.

Standard limit, not the measured value in every classroom.

Standard limit.

Theoretical design relationship, not a measured outcome for all students.

Illustrative acoustic relationship.

Health-context threshold cited by the study, not an intelligibility score.

Television speech intelligibility, captions and hearing aids

  • In a television study, 15 adults aged 59–82 with bilateral sensorineural hearing loss completed repeated speech-recognition tests (published study; older adults with hearing loss). (The benefits of hearing aids…)
  • Among young normal-hearing pilot listeners, unc captioned televised speech lists ranged from 86% to 98% correct, with a 92% mean (pilot testing; young normal-hearing listeners). (The benefits of hearing aids…)
  • Among the same pilot listeners, caption-only lists ranged from 95% to 100% correct, with a 98% mean (pilot testing; young normal-hearing listeners). (The benefits of hearing aids…)
  • The older participants’ average aided monosyllabic-word score was 56%, ranging from 6% to 96% (published study; older adults with hearing loss). (The benefits of hearing aids…)
  • 73% of older participants regularly wore their hearing aids while watching television (published study; older adults with hearing loss). (The benefits of hearing aids…)
  • 13% of older participants had ever used closed captioning in the study questionnaire (published study; older adults with hearing loss). (The benefits of hearing aids…)

Small controlled sample.

Pilot control group, not the older hearing-loss sample.

Captions were shown with sound off; pilot control group.

Standard word test, not the televised-speech outcome.

Self-reported behavior in a small sample.

Hearing loss populations linked to speech understanding

  • WHO estimates that nearly 2.5 billion people will have some degree of hearing loss by 2050 (2050 projection). (Deafness and hearing loss)
  • WHO projects that more than 700 million people will require hearing rehabilitation by 2050 (2050 projection). (Deafness and hearing loss)
  • WHO reports about 95.1 million children aged 5–19 living with hearing loss worldwide (WHO fact-sheet estimate; global, ages 5–19). (Deafness and hearing loss)
  • WHO reports 430 million people, or more than 5% of the world population, require rehabilitation for disabling hearing loss, including 34 million children (WHO fact-sheet estimate). (Deafness and hearing loss)
  • WHO defines disabling hearing loss as greater than 35 dB in the better-hearing ear (WHO definition; global definition). (Deafness and hearing loss)
  • Nearly 80% of people with disabling hearing loss live in low- and middle-income countries, according to WHO (WHO fact-sheet estimate; global by income group). (Deafness and hearing loss)
  • More than 25% of people older than 60 are affected by disabling hearing loss, according to WHO (WHO fact-sheet estimate). (Deafness and hearing loss)
  • WHO says nearly 60% of hearing loss in children is due to avoidable causes (WHO fact-sheet estimate; global children). (Deafness and hearing loss)

Projection concerns any degree of hearing loss, not speech intelligibility specifically.

Forecast of rehabilitation need, not a direct intelligibility measure.

Hearing loss population, not a speech-test result.

Disabling hearing loss is defined by hearing threshold, not directly by intelligibility score.

Clinical population definition, not a speech-recognition percentage.

Population distribution estimate.

Age-group prevalence, not a direct speech test.

Cause attribution, not speech intelligibility.

Speech intelligibility barriers, costs and measurable outcomes

  • The Global Burden of Disease 2019 analysis estimated 1.57 billion people had hearing loss in 2019, with a 95% uncertainty interval of 1.51–1.64 billion. (Hearing loss prevalence and years…)
  • The same GBD analysis estimated 43.45 million years lived with disability from hearing loss in 2019, with a 95% uncertainty interval of 29.68–61.80 million. (Hearing loss prevalence and years…)
  • A WHO-linked global cost analysis estimated total economic costs of hearing loss above US$981 billion (2021 publication estimate). (Estimating the global costs of…)
  • In that cost model, 47% of costs were attributed to quality-of-life losses (2021 publication estimate). (Estimating the global costs of…)
  • The model attributed 32% of hearing-loss costs to additional poor-health costs among people with hearing loss (2021 publication estimate). (Estimating the global costs of…)
  • The model estimated that 57% of hearing-loss costs occurred outside high-income countries (2021 publication estimate; global by country income group). (Estimating the global costs of…)
  • The model estimated that 6.5% of hearing-loss costs were for children aged 0–14 (2021 publication estimate; global, ages 0–14). (Estimating the global costs of…)

GBD hearing-loss definition and uncertainty interval; not a direct intelligibility test.

Burden model, not a speech-recognition outcome.

Modeled economic estimate, not a direct cost of poor speech intelligibility.

Model allocation, not observed household spending.

Model allocation.

Model allocation and country-income classification.

Modeled age allocation.

Table of contents