Search

This paper presents the Coswara dataset, a dataset containing diverse set of respiratory sounds and rich meta-data, recorded between April-2020 and February-2022 from 2635 individuals (1819 SARS-CoV-2 negative, 674 positive, and 142 recovered subjects). The respiratory sounds contained nine sound categories associated with variants of breathing, cough and speech. The rich metadata contained demographic information associated with age, gender and geographic location, as well as the health information relating to the symptoms, pre-existing respiratory ailments, comorbidity and SARS-CoV-2 test status. Our study is the first of its kind to manually annotate the audio quality of the entire dataset (amounting to 65 hours) through manual listening. The paper summarizes the data collection procedure, demographic, symptoms and audio data information. A COVID-19 classifier based on bi-directional long short-term (BLSTM) architecture, is trained and evaluated on the different population sub-groups contained in the dataset to understand the bias/fairness of the model. This enabled the analysis of the impact of gender, geographic location, date of recording, and language proficiency on the COVID-19 detection performance. [ABSTRACT FROM AUTHOR]

Introduction: In severe TMJ ankylosis cases, the lack of growth of the mandible creates an anatomically narrow airway with a reduced pharyngeal airway space [PAS] which predisposes these patients towards obstructive apnoea [OSA]. There is evidence in the literature that such patients experience severe discomfort during physiotherapy if such airway abnormalities are not corrected prior to ankylosis release. This eventually leads to non-compliance towards physiotherapy and increases the risk of re-ankylosis.Objective: In our study, pre-arthroplastic mandibular distraction osteogenesis [DO] was used to increase the PAS and resolve the underlying OSA prior to releasing the ankylosis.Materials and methods: Twenty-five cases of TMJ ankylosis with micrognathia and OSA were included in this prospective observational sleep study. They were further divided into a paediatric group [14 subjects] and an adult group [11 subjects]. All cases presented with a history of onset of ankylosis during childhood [before the completion of craniofacial growth] as result of which there was a lack of forward growth of the mandible. Subjects included in our study underwent initial DO of the mandible followed by a second procedure for distractor removal and ankylosis release. Questionnaires, lateral cephalograms and sleep studies were taken pre-operatively (T0), immediate post-distraction to the desired length (T1) and 12 months post the distractor removal and ankylosis release (T2). The parameters studied were PAS width, apnoea hypopnea index [AHI], O2 saturation, mouth opening and mandibular advancement.Results: The paediatric group variables were as follows: mean PAS width which increased from 3.5 mm [T0] to 9 mm [T2], mean AHI which decreased from 48.04 [T0] to 3.60 [T2], mouth opening which increased from 4.5 mm [T0] to 34 mm [T2] and mean O2 saturation which increased from 89.86% [T1] to 96.88% [T2]. The adult group variables were as follows: mean PAS width which increased from 5 mm [T0] to 11 mm [T2], mean AHI which decreased from 31.45 [T0] to 1.43 [T2], mouth opening which increased from 5 mm [T0] to 34 mm [T2] and mean O2 saturation which increased from 92.01% [T0] to 96.84% [T2]. Statistical analysis revealed that DO of the mandible significantly improved OSA by increasing the PAS which was evident by the lower AHI score. Mouth opening was also significantly improved post ankylosis release and maintained at the T2 interval. Ten subjects followed up beyond the T2 interval [mean 28 months post ankylosis release] and their data also revealed positive compliance towards physiotherapy, adequate mouth opening and maintenance of normal AHI.Conclusion: Pre-arthroplastic mandibular DO has proved to be a successful modality for treatment of OSA in TMJ ankylosis patients with stable results at 12 months. By resolving the narrow airway and OSA, compliance towards physiotherapy was improved thus reducing the risk of re-ankylosis in the long term. [ABSTRACT FROM AUTHOR]

We revisit an original concept of speech coding in which the signal is separated into the carrier modulated by the signal envelope. A recently developed technique, called frequency-domain linear prediction (FDLP), is applied for the efficient estimation of the envelope. The processing in the temporal domain allows for a straightforward emulation of the forward temporal masking. This, combined with an efficient nonuniform sub-band decomposition and application of noise shaping in spectral domain instead of temporal domain (a technique to suppress artifacts in tonal audio signals), yields a codec that does not rely on the linear speech production model but rather uses well-accepted concept of frequency-selective auditory perception. As such, the codec is not only specific for coding speech but also well suited for coding other important acoustic signals such as music and mixed content. The quality of the proposed codec at 66 kbps is evaluated using objective and subjective quality assessments. The evaluation indicates competitive performance with the MPEG codecs operating at similar bit rates. [ABSTRACT FROM AUTHOR]