A conversational speech-in-noise test administered without human supervision is desirable to reduce the need to attend the clinic. This study aimed to create and evaluate a novel question-and-answer-based speech-in-noise test called 'CHAT', where stimuli were synthesised using text-to-speech (TTS), responses captured using automatic speech recognition (ASR), and responses evaluated using a large language model...
✦ The floor
Discussion
Signed responses from readers of the wire.
Clinicians should monitor this technology; if validated in diverse populations, an LLM-driven unsupervised speech-in-noise test could extend screening access, but it is not yet ready to replace standard supervised testing.
An AI-driven, unsupervised speech-in-noise test could dramatically lower barriers to hearing screening, particularly in underserved or remote settings.
- 01An LLM-based conversational interface was used to administer a speech-in-noise hearing test without clinical supervision.
- 02Published in Trends in Hearing, a peer-reviewed audiology journal.
- 03Automated Q&A format could enable scalable remote hearing screening.
- 04Validity and reliability compared to standard clinical tests requires further study.
- 05Represents a novel application of generative AI to audiological assessment.
An LLM-driven conversational hearing test can assess speech-in-noise perception without clinical supervision.
studypartially supportedThe automated system is designed to operate without a clinician present.
studysupported- PMID
- 42858185
- DOI
- 10.1177/23312165261479670.
- Journal
- Trends in Hearing
- Publication type
- research_article
- Evidence level
- 2b
- Population
- Not fully stated; participants undergoing automated speech-in-noise hearing testing
- Intervention
- LLM-driven automated question-and-answer speech-in-noise hearing test
Primary outcomes
Feasibility and performance of automated LLM-driven speech-in-noise testing; Agreement with or equivalence to supervised clinical testing