In this paper, we present a speaker-independent dysarthric speech recognition system, with a focus on evaluating the recently released Speech Accessibility Project (SAP-1005) dataset, which includes speech data from individuals with Parkinson’s disease (PD). Despite the growing body of research in dysarthric speech recognition, many existing systems are speaker-dependent and adaptive, limiting their generalizability across different speakers and etiologies. Our primary objective is to develop a robust speaker-independent model capable of accurately recognizing dysarthric speech, irrespective of the speaker. Additionally, as a secondary objective, we aim to test the cross-etiology performance of our model by evaluating it on the TORGO dataset, which contains speech samples from individuals with cerebral palsy (CP) and amyotrophic lateral sclerosis (ALS). By leveraging the Whisper model, our speaker-independent system achieved a CER of 6.99% and a WER of 10.71% on the SAP-1005 dataset. Further, in cross-etiology settings, we achieved a CER of 25.08% and a WER of 39.56% on the TORGO dataset. These results highlight the potential of our approach to generalize across unseen speakers and different etiologies of dysarthria.

Automatic Speech Recognition (ASR) holds immense potential to provide an effective interface for assistive technologies, but its performance remains unsatisfactory for people with speech impairments such as dysarthria. Existing ASR systems struggle to accurately recognize dysarthric speech due to the significant speaker variability in dysarthric speech and the scarcity of dysarthric datasets. In this study, we propose a two-phase adaptation pipeline based on the Conformer architecture that leverages typical speech to transfer to individualized ASR models for dysarthric speakers. ASR performance is evaluated for isolated words and continuous sentences, yielding an average Word Error Rate of 21.5% on the UASpeech dataset and 12.7% on the TORGO dataset. Selectively freezing decoder layers was more often successful than selectively freezing encoder layers, suggesting that optimal performance is achieved by focusing the adaptation on the acoustic information contained in the encoder.

User feedback on software usage is utilised by developers to improve their software. Software product forums are platforms rich in software-related user feedback, such as forum threads containing bug reports or requests for new features. However, previous studies have mainly focused on analysing user feedback from software product forums as individual sentences, which can lead to missing insights and a lack of understanding of the overall context of forum posts. To fill this gap in research, this work examines user feedback found in software product forum posts to investigate the differences between content classifications found in forum sentences and posts. We manually evaluated software product forum posts collected from two open-sourced software product forums and discovered five new types of user feedback that can only be identified when examining user feedback in the form of forum posts. Additionally, we examined the association between sentence classifications found within software product forums. Our results indicate that contextual information complimenting product improvement insights can be found in software product forums, with a confidence of 0.75 and 0.69 for the association between apparent bug and application usage sentences. This information can be used to reduce manual efforts required to chase up missing contextual information when attempting to understand or fix software issues. We also provide insights into the progression of posts in software product forums at the thread-level, and our progression flowchart can be used to summarise the sequence of events in software product forum threads. Our findings reveal the importance of looking at user feedback within software product forums in the format of forum posts to identify new insights on user feedback for software improvements. Editor’s note: Open Science material was validated by the Journal of Systems and Software Open Science Board.

Abstract Speaker Identification (SI) is the task of identifying an unknown speaker of an utterance by comparing the voice biometrics of the unknown speaker with previously stored and known speaker models. Although deep learning algorithms have been successful in different speech and speaker recognition systems, they are computationally expensive and require considerable run-time resources. This paper approaches this issue by proposing an optimized text-independent SI system based on convolutional neural networks (CNNs) that not only delivers accuracies on par with state-of-the-art benchmarks but also demands significantly fewer trainable parameters. The proposed system integrates an Enhanced Multi-Active Learner framework, which distributes the complexity of the learning task among an array of learners, with a novel SI approach in which speakers are identified based on a single sound segment of voice biometrics. Here, experiments were conducted with all 1881 VoxCeleb 1 and TIMIT speakers, and results were compared with the SI systems reported in the literature that were assessed on the same speakers’ data. Results indicate that first, the proposed system outperformed the benchmark systems’ performances by delivering up to 2.43% better top-1 accuracy, and second, it reduced the number of deep learning trainable parameters by up to 95%. The proposed SI could bring offline, large-scale speaker identification to low-end computing machines without specific deep learning hardware and make the technology more affordable.

Affecting approximately 1 in 60 individuals, Autism spectrum disorder (ASD) is a lifelong developmental condition associated with substantial healthcare costs and time-consuming diagnosis, whereas early detection of ASD traits may help limit further development of the condition. To help speed up ASD diagnosis, screening approaches have been introduced, but they are not widely used due to their reliability issues since these approaches use basic scoring functions to identify traits of autism rather than intelligently learning autistic indicators from historical cases and controls. Autism Artificial Intelligence is one such system that visualizes autistic behavioral indicators and leverages a Convolutional Neural Network to learn the behavioral patterns associated with autism using historical data collected from more than 6000 ASD cases and control. Autism AI has been operational since August 2018, and more than 10,000 individuals have benefited from this platform. In this paper, after briefly explaining the system, we will analyze how well it has performed compared to AQ-10 and Q-CHAT-10 screening methods and benchmark it against the diagnosis data we have collected.

The growing application of DL makes detecting and fixing defective DL programs of paramount importance. Recent studies on DL defects report that TensorFlow API misuses represent a common class of DL defects. However to effectively detect, fix, and verify them remains an understudied problem. This paper presents the TensorFlow API misuses Detector And Fixer (TADAF) technique, which relies on 11 common API misuses patterns and corresponding fixes that we extracted from StackOverftow. TADAF statically analyses a TensorFlow program for identifying matches of any of the 11 patterns. If it finds a match, it automatically generates a fixed version of the program. To verify that the misuse brings a tangible negative effect, TADAF reports functional, accuracy, or efficiency differences when training and testing (with the same data) the original and fixed versions of the program. Our preliminary evaluation on five GitHub projects shows that TADAF detected and fixed all the API misuses.

As applications of neural networks increase in our daily lives, their practicality and accuracy become more of a challenge as they are applied to approximate more complicated functions typically composed of different dependent or independent views. While the complexity of the functions and the number of views to be approximated or simulated increases, the task becomes more complicated and more difficult in that it may eventually jeopardise the classifier’s accuracy and make the results unreliable. This paper surveys an improved active learning method called Enhanced Multi-Learner (EML) to facilitate the approximation or simulation of complex functions via neural networks by distributing the complexities of the task under simulation among an array of learners where each network is responsible for learning a specific view. We experimented with EML realisations through neural networks to solve complex problems where traditional methods did not provide adequate results. These experimental studies were conducted in three different domains and are summarised here. Legacy solutions were also provided in each experiment, and the results were compared. The experimental results indicate the superiority of EML base neural networks in dealing with sophisticated pattern recognition problems.

Software testing has been used to find software faults in order to improve its quality. To verify the software behavior, testers require test oracle. Test oracle is a reliable source of expected software behavior that provides outputs for any input specified in the software specifications and a comparator to verify actual results. While test automation requires automated oracle support, oracle automation is considered as a challenging task. These challenges are from the automation required in expected output generation and results verification. This paper presents oracle activities and the challenges to prepare automated oracle. Then a comparative study of existing automated oracle and expected output generation methods is provided. Finally, a classification of these methods is suggested based on how these methods provide automated test oracle and the tool they used. The classification explains which oracle activities will be automated by the proposed approaches.