A sterile environment in the medical field represents a crucial point for limiting the spread of bacteria and viruses both towards patients being treated and towards healthcare professionals. The pandemic faced in recent years has highlighted the benefits of contactless voice control with devices. Given the highly topical nature of this issue, we wanted to develop a project that envisaged the integration and development of a system that could allow voice control of a medical device. In order complete the specific dates, in a first phase we wanted to deepen the study of phonetics in the field of frequency. It was decided to pay attention to the natural filters present inside the human body and to the decoding of the frequency bands through the use of Automatic Speech Recognition. Subsequently, an external software was chosen for the creation of models for the implementation of the voice commands required for medical development. A second part of the project considered the analysis of the source code associated with the electric board and the deepening of the main tasks that represent the different steps for speech recognition. This phase has assumed an essential role in understanding where to insert the customized ad hoc modifications for the product. The last phase involved integrating and developing communication with the medical washer-disinfector. The testing and validation process played a key role in being able to calibrate the parameters of the speech recognition software in an optimal way.
Un ambiente sterile in ambito medico rappresenta un punto cruciale per limitare la diffusione di batteri e virus sia verso i pazienti in cura e sia verso gli operatori sanitari. La pandemia affrontata negli ultimi anni ha messo in evidenza i vantaggi del controllo vocale senza contatto con i dispositivi. Visto la forte attualità di questa tematica si è voluto sviluppare un progetto che ha previsto l'integrazione e lo sviluppo di un sistema che potesse permettere un controllo vocale su una apparecchiatura medica. Per portare a termine le specifiche date, in una prima fase si è voluto approfondire lo studio della fonetica nel campo della frequenza. Si è deciso di dedicare attenzione ai filtri naturali presenti all'interno del corpo umano e alla decodifica delle bande di frequenza tramite l'uso dell'Automatic Speech Recognition. Successivamente si è scelto un software esterno per la realizzazione di modelli per l'implementazione dei comandi vocali richiesti per lo sviluppo medico. Una seconda parte del progetto ha considerato l'analisi del codice sorgente associato alla scheda e l'approfondimento dei task principali che rappresentano i diversi step per il riconoscimento vocale. Questa fase ha assunto un ruolo essenziale per capire dove inserire le modifiche personalizzate ad hoc per il prodotto. L'ultima fase ha previsto l'integrazione e lo sviluppo di una comunicazione con il termo-disinfettore medico. Il processo di test e convalidazione ha rivestito un ruolo chiave per poter calibrare i parametri del software per il riconoscimento vocale in modo ottimale.
Integrazione e sviluppo di una scheda per il controllo vocale su una apparecchiatura di termo-disinfezione medicale.
BOVI, MIRKO
2022/2023
Abstract
A sterile environment in the medical field represents a crucial point for limiting the spread of bacteria and viruses both towards patients being treated and towards healthcare professionals. The pandemic faced in recent years has highlighted the benefits of contactless voice control with devices. Given the highly topical nature of this issue, we wanted to develop a project that envisaged the integration and development of a system that could allow voice control of a medical device. In order complete the specific dates, in a first phase we wanted to deepen the study of phonetics in the field of frequency. It was decided to pay attention to the natural filters present inside the human body and to the decoding of the frequency bands through the use of Automatic Speech Recognition. Subsequently, an external software was chosen for the creation of models for the implementation of the voice commands required for medical development. A second part of the project considered the analysis of the source code associated with the electric board and the deepening of the main tasks that represent the different steps for speech recognition. This phase has assumed an essential role in understanding where to insert the customized ad hoc modifications for the product. The last phase involved integrating and developing communication with the medical washer-disinfector. The testing and validation process played a key role in being able to calibrate the parameters of the speech recognition software in an optimal way.File | Dimensione | Formato | |
---|---|---|---|
BOVI_MIRKO.pdf
accesso riservato
Dimensione
2.91 MB
Formato
Adobe PDF
|
2.91 MB | Adobe PDF |
The text of this website © Università degli studi di Padova. Full Text are published under a non-exclusive license. Metadata are under a CC0 License
https://hdl.handle.net/20.500.12608/43332