Summary
Compared sound sent into cochlear implant processors through a telecoil and neckloop, Bluetooth and two remote-microphone accessories with sound from a loudspeaker, and tested speech in noise through each route with normal-hearing listeners.
Key findings
- Speech-shaped noise was sent from a phone, or a laptop for the neckloop route, to free-standing processors (Cochlear Nucleus 7, Advanced Bionics Naída Marvel, MED-EL Sonnet 2) by loudspeaker, Bluetooth, a Cochlear Mini Mic 2+, a Phonak Roger Select with Roger 20 receiver, and a telecoil with a Phonak Roger Neckloop (Nucleus 7 only), with the processor microphones disabled for each wireless route except MED-EL AudioStream, whose mix could not be adjusted; the output was recorded through each maker's listening-check accessory.
- On the Nucleus 7, the differences from loudspeaker presentation were generally under 5 dB above 500 Hz and larger in some lower third-octave bands; the authors attribute them mainly to the narrower bandwidth of wireless streaming, whose codecs limit the upper frequency to about 7-7.3 kHz.
- Twenty normal-hearing adults took the digits-in-noise test through headphones fed from the processors; speech reception thresholds did not differ significantly between the five routes on the Nucleus 7, or between Bluetooth and loudspeaker for any of the three brands.
- The authors note that the neckloop's placement significantly affected the volume and that the telecoil route is sensitive to magnetic interference and took more effort to set up; they recommend digital streaming for remote testing.
- For the MED-EL processor, Bluetooth needed an AudioStream adapter, which does not let the user change the mixing ratio or switch off the processor microphone.
- In Experiment 1, Results (printed page 316, Figure 1), the Nucleus 7 third-octave spectra are aligned at 1 kHz; the authors state that absolute levels are arbitrary. The reported differences therefore compare spectral shape after alignment, not absolute output. Experiment 1, Methods, explains that the AudioStream microphone remained active in a soundproof booth with negligible expected background noise.
- Experiment 1, Methods (printed page 316), says the Roger Neckloop could not connect to the smartphone, so an Apple MacBook Air played the same audio. A check using the Roger Select found no output difference between phone and laptop presentation.
- Experiment 1, Methods (printed page 315), describes listening-check outputs feeding a Samson Servo 120a amplifier and Sennheiser HDA 200 headphones on a Bruel & Kjaer 4153 artificial ear connected to a 2250 sound level meter. The setup was calibrated to 65 dBA for each condition; that is an acoustic output calibration, not a neckloop magnetic-field measurement.
- Experiment 1, Methods (printed pages 315-316), says optional processor preprocessing was disabled. The Roger Neckloop was suspended around the Nucleus 7 processor in telecoil mode, with its microphone disabled. Table 1 assigns the telecoil condition only to Nucleus 7; the article does not report repeat units for that condition.
Limits
Tested with normal-hearing listeners hearing the processors' monitor output, not with implant users, which the authors say cannot be fully generalised; a laboratory set-up with personal devices, not a venue system, and the telecoil route tested on one processor only. Published online on 5 August 2024, open access, in the April 2025 issue.
Read the original
doi.org/10.1080/14992027.2024.2382201
Cited in
Cite this source
Rootlieb JM, Polspoel S, Brienesse P, Smits C. Remote self-testing for adult cochlear implant users: the effect of wireless streaming on speech recognition in noise. International Journal of Audiology 64(4):314-319 (2025). 5 August 2024. https://doi.org/10.1080/14992027.2024.2382201 (accessed 13 October 2026).
Every finding here cites its source, and who funded each source is labelled. This site does not recommend products. Found an error or a missing study? Suggest a correction.