TRAIN ANNOUNCEMENTS
SPEECH-TO-TEXT
THE CHALLENGE: Addressing Audio AnnouncementS on Trains
More and more passengers are not able to hear audio announcements on trains. They might have missed the announcement due to hearing problems, distractions, or background noise. This situation adds to the stress passengers experience while traveling and burdens train personnel more.
WHAT IS SPEECH-TO-TEXT?
Televic GSP provides a live Speech-to-text (STT) solution for train announcements. It converts real-time audio announcements that train personnel make into text. Through the use of an API (Application Programming Interface), the transcription can be seamlessly integrated into other onboard systems for diverse applications including passenger-oriented mobile apps.
The high-quality audio recorded by our public announcement system is transcribed with a local state-of-the-art STT model. Using advanced AI models, Televic GSP's system ensures highly accurate transcriptions by incorporating specific train-related vocabulary like station names, routes, and branding.
This technology creates almost perfect subtitles, making it easier for passengers to understand live announcements. It improves communication and enhances the travel experience for everyone on board.
Benefits of speech-to-text
1. Improved passenger information
Passengers who have trouble understanding the live announcements made by train personnel can read the message simultaneously on the train screens.
2. Accuracy
Televic GSP uses the context available in the Passenger Information System (station names, branding, delays) to correct the transcription, ensuring the announcements are nearly perfectly transcribed.
3. Maximal flexibility
The transcription can be shown in real time, or the audio announcement can be played back in conjunction with the transcription. The visualization of the subtitles can be customized to the customer’s needs.
4. Maximal reliability
By using a local state-of-the-art STT model, there is no need for a stable internet connection.