Why do Cochlear Implant Users Struggle to Understand Speech in Echoey Spaces?

Prajna BK – prajnab2@illinois.edu

University of Illinois Urbana-Champaign, Speech and Hearing Science, Champaign, IL, 61820, United States

Justin Aronoff

Popular version of 2pSPb4 – Impact of Cochlear Implant Processing on Acoustic Cues Critical for Room Adaptation
Presented at the 188th ASA Meeting
Read the abstract at https://eppro01.ativ.me//web/index.php?page=Session&project=ASAICA25&id=3867053

–The research described in this Acoustics Lay Language Paper may not have yet been peer reviewed–

Have you ever wondered how we manage to understand someone in echoey and noisy spaces? For people using cochlear implants, understanding speech in these environments is especially difficult—and our research aims to explore why.

Figure 1. Spectrogram of reverberant speech before (top) and after (bottom) Cochlear Implant processing

 

When sound is produced in a room, it reflects off surfaces and lingers—creating reverberation. Reflections of both target speech and background noise make understanding speech even more difficult. However, for listeners with typical hearing, the brain quickly adapts to these reflections through short-term exposure, helping separate the speech signal from the room’s acoustic “fingerprint.” This process, known as adaptation, relies on specific sound features: the reverberation tail (the lingering energy after the speech stops), reduced modulation depth (how much the amplitude of the speech varies), and increased energy at low frequencies. Together, these cues create temporal and spectral patterns that the brain can group as separate from the speech itself.

While typical-hearing listeners adapt, many cochlear implant (CI) users report extreme difficulty understanding speech in everyday places like restaurants, where background noise and sound reflections are common. Although cochlear implants have been remarkably effective in restoring access to sound and speech for people with profound hearing loss, they still fall short in complex acoustic environments. This study explores the nature of distortions introduced by cochlear implants to key acoustic cues that listeners with typical hearing use to adapt to reverberant rooms.

The study examined how cochlear implant signal processing affects these cues by analysing room impulse response signals before and after simulated CI processing. Two key parameters were manipulated: the input dynamic range (IDR), which determines how much of the incoming sound is preserved before compression and affects how soft and loud sounds are balanced in the delivered electric signal. The second parameter, the Logarithmic Growth Function (LGF), controls how sharply the sound is compressed at higher levels. A lower LGF results in more abrupt shifts in volume, which can distort fine details in the sound.

The results show that cochlear implant processing significantly alters the acoustic cues that support adaptation. Specifically, it reduces the fidelity with which modulations are preserved, shortens the reverberation tail, and diminishes the low-frequency energy typically added by reflections. Overall, this degrades the speech clarity index of the sound, which can contribute to CI users’ difficulty communicating in reflective spaces.

Further, increasing the IDR extended the reverberation tail but also reduced the clarity index by increasing the relative contribution of reverberant energy to the total energy. Similarly, lowering the LGF factor caused more abrupt energy changes in the reverberation tail, degrading modulation fidelity. Interestingly, it also led to a more gradual drop-off in low-frequency energy—highlighting a complex trade-off.

Together, these findings suggest that cochlear implant users may struggle in reverberant environments not only because of reflections but also because their devices alter or distort the acoustic regularities that enable room adaptation. Improving how cochlear implants encode these features could make speech more intelligible in real-world, echo-filled spaces.

Myth busted: classroom acoustics can be easy and cheap

Coralie van Reenen – cvreenen@csir.co.za

Council for Scientific and Industrial Research, Council for Scientific and Industrial Research, Gauteng, 0001, South Africa

Popular version of 3pAAb – Classroom acoustics: a case study of the cost-benefit of retrofitted interventions
Presented at the 185th ASA Meeting
Read the abstract at https://doi.org/10.1121/10.0023323

Please keep in mind that the research described in this Lay Language Paper may not have yet been peer reviewed.

Achieving the right acoustic conditions for classrooms is often dismissed by school planners as being too difficult or too expensive. This is to the detriment of students who are unable to hear the teacher properly, especially for children who are being taught in their second language, as is common in South Africa. This study proves that acoustic treatment need not be difficult or costly to achieve.

To refute the notion that acoustic improvements are expensive and specialized, this experimental case study was designed and carried out in a typical classroom in the small rural village of Cofimvaba in the Eastern Cape, South Africa. The ideal classroom environment has a low ambient noise level of 35 dB and a reverberation time below 0.7 seconds, but this classroom has a reverberation time of 1 second. Reverberation time refers to the time it takes for a noise to die down and essentially refers to how much a room echoes, which negatively affects speech clarity. The experimental intervention simulated the installation of floating ceiling islands by installing different materials on the roof of temporary gazebos in the classroom.

The four materials used were acoustic ceiling tiles which represent a typical solution and three DIY solutions using carboard egg cartons, thermal insulation batting, and sponge foam bed mattresses. Each material provided an improved reverberation time. The best performing was the sponge at 0.6 seconds, while the other three materials performed equally at 0.8 seconds.

The cost of each material was reduced to a rate per square meter. The most expensive material was the acoustic ceiling tiles at R 363.85/m2 while the cheapest was the egg cartons at R 22.22/m2, or less if they are available as waste items.

The availability of materials was evaluated in terms of the distance to supply and whether the product is available in a retail store or requires a special order and delivery. The batting is available from hardware stores nationwide and could be purchased by walk-in from the local hardware store, within a 2 km radius of the site. The egg cartons could be ordered online and delivered from a packaging company within a 150 km radius. The foam mattresses could be purchased by walk-in at a local retailer within a 5 km radius of the site. The acoustic ceiling tiles were ordered online and delivered from the warehouse within a 700 km radius of the site.

Using the weighted sum model and assigning equal weighting to each attribute of acoustic performance, cost, distance to supply, and walk-in availability, a performance score for each intervention material was calculated. The batting ranked number one, followed in order by the sponge, egg cartons and lastly acoustic tiles.

The case study demonstrates that an improvement in acoustic conditions of at least a 0.2 second reduction in reverberation time can be achieved without significant cost. Although the batting did not achieve the ideal reverberation time, when only the speech frequencies were considered, it fell within the recommended maximum of 0.7 seconds.

The recommended design intervention is a frame containing batting covered with a taught fabric and suspended from ceiling hooks, thus avoiding disruptive construction works. This shows that improved classroom acoustics can be achieved without high cost or technical difficulty.

Read more: Classroom acoustics: a case study of the cost-benefit of retrofitted interventions