2022 IEEE Spoken Language Technology Workshop (SLT), Год журнала: 2024, Номер unknown, С. 766 - 773
Опубликована: Дек. 2, 2024
Язык: Английский
Heliyon, Год журнала: 2025, Номер 11(2), С. e42083 - e42083
Опубликована: Янв. 1, 2025
Sentence stimuli pervade psycholinguistics research. Yet, limited attention has been paid to the automatic construction of sentence stimuli. Given their linguistic capabilities, this study investigated efficacy ChatGPT in generating and AI tools producing auditory In three psycholinguistic experiments, examined acceptability validity AI-formulated sentences written one two languages: English Arabic. Experiment 1 3, participants gave AI-generated similar or higher ratings than human-composed 2, Arabic received lower counterparts. The AI-developed relied on design, with only 2 demonstrating target effect. These results highlight promising role as a developer, which could facilitate research increase its diversity. Implications for were discussed.
Язык: Английский
Процитировано
0
Nippon Onkyo Gakkaishi/Acoustical science and technology/Nihon Onkyo Gakkaishi, Год журнала: 2024, Номер unknown
Опубликована: Янв. 1, 2024
Язык: Английский
Процитировано
1
The Journal of the Acoustical Society of America, Год журнала: 2024, Номер 156(5), С. 3574 - 3587
Опубликована: Ноя. 1, 2024
Because a reference signal is often unavailable in real-world scenarios, reference-free speech quality and intelligibility assessment models are important for many processing applications. Despite great number of deep-learning that have been applied to build non-intrusive approaches achieve promising performance, studies focusing on the hearing impaired (HI) subjects limited. This paper presents HASA-Net+, multi-objective hearing-aid model, building upon our previous work, HASA-Net. HASA-Net+ improves HASA-Net several ways: (1) inclusivity both normal-hearing HI listeners, (2) integration with pre-trained foundation fine-tuning techniques, (3) expansion predictive capabilities cover diverse conditions, including noisy, denoised, reverberant, dereverberated, vocoded speech, thereby evaluating its robustness, (4) validation generalization capability using an out-of-domain dataset.
Язык: Английский
Процитировано
1
2022 IEEE Spoken Language Technology Workshop (SLT), Год журнала: 2024, Номер unknown, С. 803 - 810
Опубликована: Дек. 2, 2024
Язык: Английский
Процитировано
1
Опубликована: Янв. 1, 2024
Язык: Английский
Процитировано
0
Опубликована: Ноя. 17, 2024
Evaluating Text-To-Speech (TTS) systems is challenging, as the increasing quality of synthesis makes it difficult to discriminate models’ ability reproduce prosodic attributes, especially for Brazilian Portuguese. Offline evaluation metrics do not capture our genuine reactions audio stimuli. Therefore, we propose an online method using eye-tracking. Our experiments with 76 annotators show a reasonable correlation between EyetrackingMOS and MOS, well reduction in total time. We believe this metric provides precise potentially fast information complement existing methods.
Язык: Английский
Процитировано
0
Computer Speech & Language, Год журнала: 2024, Номер 90, С. 101747 - 101747
Опубликована: Ноя. 8, 2024
Язык: Английский
Процитировано
0
2022 IEEE Spoken Language Technology Workshop (SLT), Год журнала: 2024, Номер unknown, С. 766 - 773
Опубликована: Дек. 2, 2024
Язык: Английский
Процитировано
0