Dysarthria is a motor speech disorder that results in slow and often incomprehensible speech. Speech intelligibility significantly impacts communication, leading to barriers in social interactions. Dysarthria is often a characteristic of neurological diseases including Parkinson's and ALS, yet current tools lack generalizability across languages and levels of severity. In this study, we present a unified AI-based multilingual framework that addresses six key components: (1) binary dysarthria detection, (2) severity classification, (3) clean speech generation, (4) speech-to-text conversion, (5) emotion detection, and (6) voice cloning. We analyze datasets in English, Russian, and German, using spectrogram-based visualizations and acoustic feature extraction to inform model training. Our binary detection model achieved 97% accuracy across all three languages, demonstrating strong generalization across languages. The severity classification model also reached 97% test accuracy, with interpretable results showing model attention focused on lower harmonics. Our translation pipeline, trained on paired Russian dysarthric and clean speech, reconstructed intelligible outputs with low training (0.03) and test (0.06) L1 losses. Given the limited availability of English dysarthric-clean pairs, we fine-tuned the Russian model on English data and achieved improved losses of 0.02 (train) and 0.03 (test), highlighting the promise of cross-lingual transfer learning for low-resource settings. Our speech-to-text pipeline achieved a Word Error Rate of 0.1367 after three epochs, indicating accurate transcription on dysarthric speech and enabling downstream emotion recognition and voice cloning from transcribed speech. Overall, the results and products of this study can be used to diagnose dysarthria and improve communication and understanding for patients across different languages.
Indoor radon constitutes a public health issue in various regions across the United States as the second leading cause of lung cancer following tobacco smoke. The U.S. Environmental Protection Agency advises radon mitigation interventions for residential buildings with indoor radon concentrations exceeding the threshold level of 4 pCi/L. Despite considerable research assessing the technical effectiveness of radon mitigation systems, there remains a gap in understanding their broader influence on occupant behavior and preferences in residential design. This study aims to investigate the impact of residing in radon-mitigated homes within the Commonwealth of Kentucky—an area known for elevated radon concentrations—on occupants’ preferences regarding healthy home design attributes. The objectives of this research are twofold: firstly to determine if living in radon-mitigated homes enhances occupant awareness and consequently influences their preferences toward health-related home attributes and secondly to quantitatively evaluate and compare the relative significance homeowners assign to health-related attributes such as indoor air quality, thermal comfort, and water quality relative to conventional attributes including home size, architectural style, and neighborhood quality. The overarching purpose is to explore the potential role radon mitigation initiatives may play in motivating occupants towards healthier home construction and renovation practices. Using choice-based conjoint (CBC) analysis, this paper compares preferences reported by homeowners from radon-mitigated homes against those from non-mitigated homes. While the findings suggest a relationship between radon mitigation and increased preference for indoor air quality, the cross-sectional design limits causal interpretation, and the possibility of reverse causation—where health-conscious individuals are more likely to seek mitigation—must be considered. The results provide novel insights into how radon mitigation efforts might effectively influence occupant priorities towards integrating healthier design elements in residential environments.