Demos and videos
Control of a Robosapiens robot by using voice
This project was developed by Vicent Tamarit under my supervision as final work for Computer Science B.Sc. in 2006; implements Spanish speech recognition and interpretation to send IR orders to a Robosapiens.
REVOTE: speech aided handwritten text transcription
The REVOTE project (2007-2009) involved researchers Vicent Alabau, Antonio Lagarda, José Ramón Navarro, and Enrique Vidal under my coordination to create an interface to assist handwritten text transcription by using speech recognition. You can see the behaviour of the prototype in the following Flash video (click on right green arrow to move) or its MPEG video version.
Voice-based web surfing
This project was developed by Marc Franco under my supervision as final work for Computer Science B.Sc. in 2011; implements Spanish speech recognition and interpretation (performed by iATROS) to manage Firefox web browser by using voice, including the reading of links to follow them.
Hermes: distributed speech recognition and translation system based on Android
This project was developed by Emilio Granell under the supervision of Vicent Tamarit and myself as final work for Telecommunications Degree in 2012; implements a distributed speech recognition and translation system that recognises Spanish sentence in a desk hotel domain and translates them into English; it is based on Android tablets as client and on a Debian server running an application that receives the speech signal and uses iATROS as recogniser/translator to return the recognition result to the client.First video shows the basic capabilities of Hermes
Second video shows a complete interaction
We plan to distribute Hermes in the future for demo purposes
COPS: Computer Programming using Speech
The COPS project is still in development; it is a plug-in for the Eclipse IDE that implements functionalities that allow people with functional diversity an easier programming task. The initial version was developed by Santiago Sánchez Alepuz (plug-in implementation), Natividad Prieto (data and language models) and myself (speech recognition), and allows speech recognition in Spanish for dictating Java programs.First video shows the COPS interface
Second video shows the interaction with all the features of COPS
The COPS system is also available for English dictation; this video shows the interface and features of this English version
The most recent update for the COPS system, developed by Andrés Font in his final degree work (developed under my supervision), from a refactorized code by Alison Beltrán in her final degree work (developed under the supervision of Natividad Prieto and myself), allows to dictate some prefined identifiers and menu navigation by using voice. This video shows these new features (in Spanish only).
Voice-based e-mail client
This project was developed by Pablo Pellicer under my supervision as final work for Computer Science B.Sc. in 2013; implements Spanish speech recognition and interpretation (performed by iATROS) to manage Thunderbird e-mail client by using voice, including access to menu, fields, and the most usual functionalities
Voice-based Desktop Manager
This project was developed by Cristian Fernández under my supervision as final work for Computer Science Degree in 2014; implements Spanish speech recognition and interpretation (performed by iATROS) to manage an Ubuntu desktop by using voice to employ commands related to managing windows and several applications (calculator, agenda, audio mixer, etc.).
Speranto: speech recognition and translation for Google Glass
This project was developed by Pedro Pérez Sánchez under my supervision as final work for Computer Science Degree in 2014; implements the Speranto Google Glass application, with similar functionalities to those of the Hermes system.
Percepción project demo
The Percepción project was a technology transfer project that integrated speech recognition and understanding, multimodal dialogue interaction, geolocalisation, and augmented reality to obtain locations of interests and routes to them in a restricted environment (more specifically, the Smart Campus system for Universitat Jaume I campus, was implemented). The participants were FHC25 (now SmartProjects), INIT-UJI, and PRHLT-UPV (Emilio Granell, José Miguel Benedí, Alberto Sanchis - now at VRAIN, and myself).The video, which was presented at IberSpeech 2014 Demo Sessions, presents the full speech and dialogue functionality and how augmented reality is employed after speech interaction to show the points of interests and the routes to them.
Félix: a mobile articulated robot that recognises speech commands
This project was developed by Mario Parreño Lara under my supervision as final work for Computer Science Degree in 2017; it consist of the construction of an articulated robot that accepts voice commands processed by the iATROS recognition system and interpreted by a script that sends the wifi orders to the robot.
If you are interested in any of these systems or in developing similar speech-based applications as final work for your degree, contact me by sending an e-mail to cmartine_AT_dsic.upv.es