In a fast-paced world, which must comply to the technological demands coming from an immense quantity of users, the necessary time to develop and maintain applications represents a key value inside the production cycle. The creation of re-usable interfaces, which can provide a nimble way to concentrate the management for distributed system into fewer focal points is fundamental to accomplish even the hardest of the objectives. Between the latest innovations, those which are experiencing a fast growth and an ever increasing coverage over the various final user groups are Vocal assistants. These tools, that offer as their strength the immediacy and the easiness in their user interfaces, are, on the other hand, expensive for developers in terms of required time and applications portability over different systems. In this thesis work, a solution will be advanced to the difficult problem of the multi-platform approach caused by the vocal assistants proprietary development environments and it will be discussed the efficiency of an interface built starting from the Xatkit framework, as developed by the research team from UOC - Open University of Catalunia. In particular, Amazon Alexa Skills and Google Voice Assistant Actions will be considered to compare the differences or common points in user’s input retrieval approaches and the transition of the text-to-intent-recognition in Xatkit.
In un mondo in continua e veloce trasformazione, il quale deve soddisfare le richieste tecnologiche provenienti da una notevole quantit`a di utenti, il tempo di sviluppo e manutenzione delle applicazioni deve rappresentare una delle variabili principali nei cicli di produzione. La creazione di interfacce che permettono di essere riutilizzate e allo stesso tempo di concentrare la gestione di sistemi distribuiti in pochi punti nevralgici sono fondamentali per raggiungere anche gli obiettivi pi`u complessi. Tra le ultime innovazioni, quelle con una crescita pi`u notevole e una copertura sempre maggiore di tutte le fasce di utenza sono gli assistenti vocali. Questi strumenti, che propongono come loro punto di forza l’immadiatezza e la facilit`a di utilizzo dell’interfaccia utente, sono d’altro canto dispendiosi per gli sviluppatori in termini di tempo di gestione e portabilit`a delle applicazioni. In questo lavoro di tesi si proporr`a una soluzione al problema del difficile approccio multipiattaforma, causato dagli ambienti di sviluppo propietari dei servizi di assistenti vocali e si discuter`a dell’efficienza di un’interfaccia sviluppata partendo dalla base del framework Xatkit, sviluppato dal team di ricerca dell’universit`a UOC - Universitat Oberta de Catalunya. In particolare, si far`a attenzione all’ambiente di sviluppo delle Skills di Amazon Alexa e delle Actions di Google Voice Assistant, confrontando gli approcci utilizzati per il recupero dell’input dell’utente e spostando il lavoro di riconoscimento text-to-intent in Xatkit.
Vocal assistants developing interface using Xatkit Bot platform
BARUFFINI, LUCA
2018/2019
Abstract
In a fast-paced world, which must comply to the technological demands coming from an immense quantity of users, the necessary time to develop and maintain applications represents a key value inside the production cycle. The creation of re-usable interfaces, which can provide a nimble way to concentrate the management for distributed system into fewer focal points is fundamental to accomplish even the hardest of the objectives. Between the latest innovations, those which are experiencing a fast growth and an ever increasing coverage over the various final user groups are Vocal assistants. These tools, that offer as their strength the immediacy and the easiness in their user interfaces, are, on the other hand, expensive for developers in terms of required time and applications portability over different systems. In this thesis work, a solution will be advanced to the difficult problem of the multi-platform approach caused by the vocal assistants proprietary development environments and it will be discussed the efficiency of an interface built starting from the Xatkit framework, as developed by the research team from UOC - Open University of Catalunia. In particular, Amazon Alexa Skills and Google Voice Assistant Actions will be considered to compare the differences or common points in user’s input retrieval approaches and the transition of the text-to-intent-recognition in Xatkit.File | Dimensione | Formato | |
---|---|---|---|
Luca_Baruffini___Thesis___Vocal_assistants_bots_interface.pdf
accessibile in internet per tutti
Descrizione: Testo della tesi
Dimensione
1.15 MB
Formato
Adobe PDF
|
1.15 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/152764