European Commission Joint Research Centre, Institute for the Protection and Security of the Citizen, VOICE Project
English | Italiano

Frequently Asked Questions

1) What is the VOICE Project?

2) What is NOT the VOICE Project?

3) Project's results (conferences, school lessons, telephone calls)

4) Join the VOICE-Schools Project

5) Use of the VOICE prototype in a classroom

6) What is deafness?

7) What is a coclea implant?


VOICE Project is a Support Action aimed at improving the existing information technology tools, which can bridge the gap between people with hearing impairment and the hearing world. It represents an effort to mediate the real needs of the deaf with how the service providers perceive them.

In the analysis of the new products available on the market, carried out together with the various subjects involved, the Project is focused on the development and test of a tool for the subtitling of lectures, school classes, television broadcasts and telephone calls.

The technical aspect, even if basic, is a minor focus, versus public awareness scopes. The developed prototypes are used to subtitle lectures, which can, thus, be followed by people with hearing impairments as much as by any other participant. At the same time, both the audience and the authorities can become convinced of the feasibility of ideas, which, so far, had seemed unrealisable dreams.


VOICE Project is not a commercial software, it is not a product which can be purchased by the final user ready for its use.

The purpose of the development of some prototypes was to demonstrate that a currently widespread technology (a PC, an audio card, a video camera) and a limited customisation of software tools available on the market (in this case Dragon System and IBM) allowed to create tools, which could effectively overcome communication barriers.

The demo basic version of such prototypes can be downloaded via Internet. In return, users are asked to provide information regarding the implemented applications or the encountered difficulties.

The system is easy to use, however, experimentation is recommended to expert users, who can transfer their results to a larger public of users (associations, rehabilitation centres, schools, and lectures' organisers).


Lectures' subtitling has, by this time, reached excellent levels. A speaker who pronounces distinctly, after having trained the computer to his voice and, above all, to his specific vocabulary, can automatically generate his or her interventions' subtitles almost without mistakes. In addition, the written text of the speech is immediately available for corrections and printing.

The use is extremely effective during lectures with an audience with a large number of hearing-impaired participants, because the attention of both the speaker and the audience is focused on the subtitles' visual communication channel. The speaker is therefore in the ideal condition to speak with a rhythm which allows the creation of subtitles which can be easily read (to make regular breathing pauses, to interrupt the subtitling, at the end of short sentences with a proper sense). In addition, the interest for the contents diverts the attention from possible errors and makes the possible repetition of the last words more acceptable.

The same results can be obtained with school lessons, if prepared, but there are several occasions on which the situation requires improvisation and words not known by the system. Whereas in some situations this can represent a difficulty, in other cases it can be an additional spur. For instance, during a foreign language lesson, words, which are apparently misinterpreted because with a particular sound can be highlighted.

A similar situation exists as for television broadcasts, where the journalists' professionalism would allow an excellent quality of the recognition, but the development of the relevant applications is slowed down by the risk of mistakes during a live broadcast. Therefore, future experimentation is under definition.

Telephone conversations still represent a challenge for the future. The limits of the transmission band and background noises are important. Different tests have been carried out to demonstrate the viability of the approach and to identify the difficulties, which should be the focus of relevant researches. The use of a restricted vocabulary remarkably improves the quality of recognition.

An important element is represented by the fact that the so-far available telephone devices for the deaf (Text-Telephones), that is keyboards with display, require the same specific tool for both interlocutors, thus limiting the number of people involved in the telephone communication. The use of a voice recognition system requires the installation of the specific tool, that is the computer, only in the house of the hearing impaired interlocutor, whereas his or her friends must become familiar with the system only the first time.

However, there are still psychological obstacles for those who speak at one end of the telephone line, without being able to see the subtitles generated on the computer screen in the house of the hearing impaired interlocutor. This requires an echo created by vocal synthesis, which slows down the system.


The first step is becoming familiar with voice recognition tools, for instance by using a standard product available on the market to dictate texts, and by showing its use to the student and to colleagues.

In parallel, another step is contacting the schools, which participate to the VOICE Project and sharing their experiences with the colleagues, via the Forum pages. In particular, the Arona II Education Office looks after the co-ordination of such activities and can be contacted via this Forum or directly by e-mail.


The experimentation in the classrooms highlighted that subtitles can, in many cases, be a support to a better understanding also for normal hearing students, mostly for foreign students, whose presence is the classrooms is stronger and stronger.

The tool proved to be an effective aid also for the study of foreign languages, both to support the correct pronunciation of words and sentences, and to centre the focus on badly pronounced words and on possible alternative interpretations of words with a similar sound.

The tool can be considered a teaching aid, which catches the attention of the students and provides a guidance to teachers for the preparation of effective lessons, supported also by visuals, regardless of the subject and of the situation. Basically, it is a complementary tool, which is worth knowing and using more or less frequently, as teachers already do with videotapes or Power Point slideshows.

Among the on-going experiences, the president of the Italian Association of Religion Teachers (Associazione Italiana Insegnanti di Religione, ANIR), intends to investigate the use of the tool to better catch the students' attention during the religion lessons, which, besides, are usually being repeated in a larger number of classes. To get information about such application, a contact can be established via this Forum or directly by e-mail.


Please refer to the exhaustive responses provided in the "Ascolta e Vivi Onlus" association's site, courtesy of Ing. Luca Del Bo.


top of page