Nowadays there is a tendency to use serial electronic devices for passenger informing via a set of display and audio announcements on many urban passenger transport routes. In some cases there is a need to renew the existing display and audio records in the information system, in particular on bus routes (as opposed to the Metro information system). Most frequently, these cases are related to:
- the bus lines re-routing;
- the street names changings;
- the emergence of new routes;
- placing advertisements;
- warning message announcing, e.g. “Dear passengers, please pay your travel expenses in time”.
Adding new announcements to the existing records or making new records result in extra costs and time and usually requires special studio equipment. Sometimes there appear some organizational challenges related to making records by the same speaker, which is often occasional process with considerable difference in time.
As an alternative, the Automated Workstation (AWS) for preparing audio information, which was developed by the Speech Synthesis&Recognition Laboratory of the UIIP NAS of Belarus, may be used. The AWS is constructed on the basis of a standard personal computer and special TTS software and serves for voice output of text information. Speech synthesizer itself reads any text information out loud instead of merely playing previously recorded audio files.
The main functional capabilities of TTS system include:
- the conversion of a given text into speech and creation of digital audio files in Belarusian or Russian, on the basis of the user’s choice
- the selection of a voice from the voice database for audio files creation (3 male voices, 2 female voices);
- the selection of an intonation contour from the procodic database to create audio files (4 different types);
- audio file recording to a disk, to a cassette, etc.
Moreover, the speech synthesizer makes it possible to change and customize the existing voices or to add new voices and intonation contours to the TTS database. Given the fact that our Laboratory have also developed the prototype of multi-language speech synthesizer for Slavic languages, the AWS can also be adapted to other languages.
The speech synthesizer works on any Windows computer. Minimum system requirements: 20 Mb of available hard-disk space for installation, 166 MHz CPU processor, 32 Mb of RAM, sound card.