In speech audiometry, the speech-recognition threshold (SRT) is usually established by adjusting the signal-to-noise ratio (SNR) until 50% of the words or sentences are repeated correctly. However, these conditions are rarely encountered in everyday situations. Therefore, for a group of 15 young participants with normal hearing and a group of 12 older participants with hearing impairment, speech-recognition scores were determined at SRT and at four higher SNRs using several stationary and fluctuating maskers. Participants’ verbal responses were recorded, and participants were asked to self-report their listening effort on a categorical scale (self-reported listening effort, SR-LE). The responses were analyzed using an Automatic Speech Recognizer (ASR) and compared to the results of a human examiner. An intraclass correlation coefficient of r = .993 for the agreement between their corresponding speech-recognition scores was observed. As expected, speech-recognition scores increased with increasing SNR and decreased with increasing SR-LE. However, differences between speech-recognition scores for fluctuating and stationary maskers were observed as a function of SNR, but not as a function of SR-LE. The verbal response time (VRT) and the response speech rate (RSR) of the listeners’ responses were measured using an ASR. The participants with hearing impairment showed significantly lower RSRs and higher VRTs compared to the participants with normal hearing. These differences may be attributed to differences in age, hearing, or both. With increasing SR-LE, VRT increased and RSR decreased. The results show the possibility of deriving a behavioral measure, VRT, measured directly from participants’ verbal responses during speech audiometry, as a proxy for SR-LE.
Trends in hearing Thousand Oaks, Calif. : Sage, 2014 28(2024), Seite 1-20 Online-Ressource
Hörtests mit Sprache im Störgeräusch sind ein wichtiges Diagnoseinstrument, um das Sprachverstehen eines Hörers zu erfassen. In der Sprachaudiometrie gibt es mehrere gut etablierte klinische Messverfahren, welche jedoch meist mit einem hohen Messaufwand verbunden sind, da ein Spezialist den sprachaudiometrischen Test durchführen muss. Diese Arbeit geht dieses Problem mit selbst durchgeführte Testverfahren, welche automatische Spracherkennung (ASR) nutzen, an. Hierbei werden zwei Szenarien berücksichtigt: einer gut kontrollierten Laborumgebung mit einem lokalen ASR-System, welches speziell für diesen Zweck entwickelt wurde und eine Screening-Anwendung, bei der die ASR-Komponente eines Smart Speakers verwendet wird, also eines hochwertigen, kommerziell erhältlichen Lautsprechers, der mit einem virtuellen Assistenten verbunden ist. Die beiden vorgeschlagenen Systeme werden mit insgesamt 139 Probanden evaluiert, welche ein breites Spektrum an Hörfähigkeiten abdecken: Normalhörende, leicht-, mittel- und hochgradig hörgeschädigte Personen sowie Personen mit Cochlea-Implantaten. Die Genauigkeit der automatisieren Messung ist dabei im gleichen Bereich, wie wenn die Messung mit einem menschlichen Versuchsleiter durchgeführt wird.
Speech-in-noise tests are an important tool for assessing the speech recognition ability of a listener. While several well-established clinical measurement procedures exist, most come with the drawback of a high measurement effort, since a specialist needs to conduct the speech audiometric test. This work addresses this issue by proposing self-measurement applications utilizing automatic speech recognition (ASR). Two different application scenarios are considered: a well-controlled laboratory environment with a locally running ASR system developed specifically for this purpose and a screening application using the ASR component of a smart speaker - i.e., a commercially available high-quality speaker connected to a virtual assistant. The two systems proposed are evaluated with 139 subjects in total - covering a wide range of hearing abilities: normal-hearing listeners, mildly-, moderately- and severely hearing-impaired subjects, as well as listeners with cochlear implants. The measurement accuracy with the unsupervised procedure was found to be in the same range as when conducting the test with a human supervisor.
IEEE International Conference on Acoustics, Speech and Signal Processing (44. : 2019 : Brighton) 2019 IEEE International Conference on Acoustics, Speech, and Signal Processing Piscataway, NJ : IEEE, 2019 (2019), Seite 636-640 1 Online-Ressource
Proceedings of the International Symposium on Auditory and Audiological Research 7 Ballerup : The Danavox Jubilee Foundation, 2019 (2019), Seite 373-380 1 Online-Ressource
INTERSPEECH (19. : 2018 : Hyderabad, Telangana) Speech research for emerging markets in multilingual societies ; Volume 2 of 6 Red Hook, NY : Curran Associates, Inc., 2019 (2018), Seite 976-980 Seite 636-1268
