1,720,955 research outputs found
Dataset supporting the publication "Improved tactile speech robustness to background noise with a dual-path recurrent neural network noise-reduction method"
No full textThis dataset supports the publication:
Fletcher, M., Perry, S., Thoidis, I., Verschuur, C., & Goehring, T. (2024). Improved tactile speech robustness to background noise with a dual-path recurrent neural network noise-reduction method. Scientific Reports.
This dataset contains three CSV files: one for the objective assessment of the audio, one for the objective assessment of the tactile signal, and one for the behavioural assessment.
The objective audio CSV file shows the eSTOI and SI-SDR scores for each model at each of the SNRs tested for the four different noise reduction methods and with no noise reduction. Data is shown for the Party noise and ITASS noise.
The objective tactile CSV file shows the SI-SDR scores for each model at each of the SNRs tested for the four different noise reduction methods and with no noise reduction. Rows show different sentences. Column headings stat the processing applied (no processing, log-MMSE, or DPRNN methods), the SNR, and whether the data is for the male ("M") or female ("F") talker.
The behavioural CSV file shows the participant number (matching the number used for the data presented in the published article associated with this dataset), dominant hand (left/right), wrist height, width, and circumference (mm), 31.5 Hz threshold and 125 Hz vibro-tactile detection threshold at the fingertip (ms/2), wrist temperature (0C), gender, age, and percentage correct for sentence identification in each condition. The header name for each condition shows whether or not noise reduction ("NR") was applied, the SNR, and whether the talker was male or female.
All data was collected at the University of Southampton, U.K.</span
Improved tactile speech robustness to background noise with a dual-path recurrent neural network noise-reduction method
No full textMany people with hearing loss struggle to understand speech in noisy environments, making noise robustness critical for hearing-assistive devices. Recently developed haptic hearing aids, which convert audio to vibration, can improve speech-in-noise performance for cochlear implant (CI) users and assist those unable to access hearing-assistive devices. They are typically body-worn rather than head-mounted, allowing additional space for batteries and microprocessors, and so can deploy more sophisticated noise-reduction techniques. The current study assessed whether a real-time-feasible dual-path recurrent neural network (DPRNN) can improve tactile speech-in-noise performance. Audio was converted to vibration on the wrist using a vocoder method, either with or without noise reduction. Performance was tested for speech in a multi-talker noise (recorded at a party) with a 2.5-dB signal-to-noise ratio. An objective assessment showed the DPRNN improved the scale-invariant signal-to-distortion ratio by 8.6 dB and substantially outperformed traditional noise-reduction (log-MMSE). A behavioural assessment in 16 participants showed the DPRNN improved tactile-only sentence identification in noise by 8.2%. This suggests that advanced techniques like the DPRNN could substantially improve outcomes with haptic hearing aids. Low-cost haptic devices could soon be an important supplement to hearing-assistive devices such as CIs or offer an alternative for people who cannot access CI technology
Restoring speech intelligibility in complex acoustic environments for people with hearing loss using deep learning techniques
No full textSpeech perception in complex acoustic environments, especially in conditions with background noise and multiple speakers, presents a substantial challenge for individuals with hearing loss, significantly impacting their social, psychological, and occupational lives. This dissertation focuses on leveraging deep learning for audio processing and analysis tasks, with a particular emphasis on enhancing speech perception in challenging listening conditions for individuals with hearing loss. Specifically, it aims to develop methods that leverage psychoacoustic and semantic factors to enhance speech intelligibility in noisy environments. In this context, a method utilizing deep learning techniques for single-channel speech signal processing was developed, based on the temporal coding mechanisms of the human auditory system. The proposed method showed improved objective results in the enhancement of speech in background noise. The research extends to the semantic audio analysis domain, aiming for more efficient and robust adaptation of deep learning techniques to dynamic sound scenes. By employing semantic analysis techniques for speaker verification, a method was developed to enhance the voice of a target speaker in conditions with background noise and multiple competing speakers. The results showed significant improvements in speech intelligibility in environments with background noise and multiple overlapping speakers, both for listeners with hearing loss and those with normal hearing. Our findings highlight the potential of deep learning in computational audition, offering solutions to the challenges of speech perception in complex acoustic environments. We encourage broader in-depth research into hearing technology for practical applications, aiming for viable improvements in the listening experience for people with hearing loss through the use of hearing devices.Η αντίληψη της ομιλίας σε σύνθετα ηχητικά περιβάλλοντα, ιδίως σε συνθήκες με θόρυβο και πολλαπλούς ομιλητές, αποτελεί μία από τις σημαντικότερες προκλήσεις για άτομα με απώλεια ακοής, επηρεάζοντας σημαντικά την κοινωνική, ψυχολογική και εργασιακή τους ζωή. Η παρούσα διατριβή επικεντρώνεται στην αξιοποίηση της βαθιάς μάθησης ως εργαλείου για την επεξεργασία και την ανάλυση του ήχου, με ιδιαίτερη έμφαση στην αντίληψη της ομιλίας από άτομα με απώλεια ακοής. Συγκεκριμένα, επιδιώκεται η ανάπτυξη μεθόδων που ενσωματώνουν ψυχοακουστικούς και σημασιολογικούς παράγοντες του ηχητικού περιεχομένου, με στόχο την αποκατάσταση της καταληπτότητας της ομιλίας σε σύνθετα ηχητικά περιβάλλοντα. Στο πλαίσιο αυτό, αναπτύχθηκε μία μέθοδος που αξιοποιεί τεχνικές βαθιάς μάθησης για την επεξεργασία μονοφωνικών σημάτων ομιλίας, βασιζόμενη στη χρονική κωδικοποίηση του ήχου από το ανθρώπινο ακουστικό σύστημα. Η προτεινόμενη μέθοδος έδειξε βελτιωμένα αποτελέσματα ενίσχυσης της ομιλίας σε συνθήκες με θόρυβο υποβάθρου. Η έρευνα επεκτείνεται στη σημασιολογική ανάλυση του ηχητικού περιεχομένου με στόχο την αποτελεσματικότερη και πιο εύρωστη προσαρμογή των τεχνικών βαθιάς μάθησης σε δυναμικές ηχητικές σκηνές. Με τη χρήση τεχνικών σημασιολογικής ανάλυσης για την επαλήθευση ομιλητή, αναπτύχθηκε μία μέθοδος για την ενίσχυση της φωνής ενός ομιλητή-στόχου σε συνθήκες με θόρυβο και άλλους ανταγωνιστικούς ομιλητές. Τα αποτελέσματα έδειξαν σημαντικές βελτιώσεις στην καταληπτότητα της ομιλίας, τόσο για άτομα με απώλεια ακοής όσο και για άτομα με φυσιολογική ακοή, σε περιβάλλοντα με θόρυβο υποβάθρου και πολλαπλούς ταυτόχρονους ομιλητές. Τα ευρήματα της παρούσας διατριβής καταδεικνύουν τις δυνατότητες της βαθιάς μάθησης στον τομέα της υπολογιστικής ακρόασης, προβάλλοντας λύσεις για την αντιμετώπιση των προκλήσεων στην αντίληψη της ομιλίας σε σύνθετα ηχητικά περιβάλλοντα. Ενθαρρύνεται η ευρύτερη εφαρμογή και εμβάθυνση στην έρευνα των τεχνολογιών ήχου για πρακτικές εφαρμογές, με στόχο την ουσιαστική βελτίωση της εμπειρίας ακρόασης για άτομα με απώλεια ακοής με τη χρήση ακουστικών συσκευών
Going Beyond Counting First Authors in Author Co-citation Analysis
Full text linkThe present study examines one of the fundamental aspects of author co-citation analysis (ACA) - the way co-citation
counts are defined. Co-citation counting provides the data on which all subsequent statistical analyses and mappings
are based, and we compare ACA results based on two different types of co-citation counting - the traditional type that
only counts the first one among a cited work's authors on the one hand and a non-traditional type that takes into
account the first 5 authors of a cited work on the other hand. Results indicate that the picture produced through this non-traditional author co-citation counting contains more coherent author groups and is therefore considerably clearer. However, this picture represents fewer specialties in the research field being studied than that produced through the traditional first-author co-citation counting when the same number of top-ranked authors is selected and analyzed. Reasons for these effects are discussed
Variations on the Author
Full text link“Variations on the Author” discusses two of Eduardo Coutinho’s recent films (Um Dia na Vida, from 2010, and Últimas Conversas, posthumously released in 2015) and their contribution to the general question of documentary authorship. The director’s filmography is characterized by a consistent yet self-effacing form of authorial self-inscription: Coutinho often features as an interviewer that rather than express opinions propels discourses; an interviewer that is good at listening. This mode of self-inscription characterizes him as an author who is not expressive but who is nonetheless markedly present on the screen. In Um Dia na Vida, however, Coutinho is completely absent form the image, while Últimas Conversas, on the contrary, includes a confessional prologue that moves the director from the margins to the center of his films. This article examines the ways in which these works stand out in the filmography of a director who offers new insights into the notion of cinematic authorship
Appropriate Similarity Measures for Author Cocitation Analysis
Full text linkWe provide a number of new insights into the methodological discussion about author cocitation analysis. We first argue that the use of the Pearson correlation for measuring the similarity between authors’ cocitation profiles is not very satisfactory. We then discuss what kind of similarity measures may be used as an alternative to the Pearson correlation. We consider three similarity measures in particular. One is the well-known cosine. The other two similarity measures have not been used before in the bibliometric literature. Finally, we show by means of an example that our findings have a high practical relevance.information science;Pearson correlation;cosine;similarity measure;author cocitation analysis
Dispelling the Myths Behind First-author Citation Counts
Full text linkWe conducted a full-scale evaluative citation analysis study of scholars in the XML research field to explore just how different from each other author rankings resulting from different citation counting methods actually are, and to demonstrate the capability of emerging data and tools on the Web in supporting more realistic citation counting methods. Our results contest some common arguments for the continued
use of first-author citation counts in the evaluation of scholars, such as high correlations between author rankings by first-author citation counts and other citation
counting methods, and high costs of using more realistic citation counting methods that are not well-supported by the ISI databases. It is argued that increasingly available digital full text research papers make it possible for citation analysis studies to go beyond what the ISI databases have directly supported and to employ more
sophisticated methods
koamabayili/VECTRON-author-checklist: VECTRON author checklist
No full textWe have done our best to complete the author checklist relating to the use of animals in the hut study. Note that the objective for the hut study was to evaluate the IRS treatment applications for residual efficacy against Anopheles mosquitoes, including the local An. coluzzii mosquito population. Cows were only used to attract mosquitoes into the huts and no tests were carried out directly on the cows. The author checklist is intended for use with studies where experiments are carried out on animals, which is why we have had such difficulty in completing this for the hut study, as many of the questions do not relate to how the cows were used
- …
