Respeaking is a voice transcription technique to produce real-time subtitles of live TV programs but also of court reports and real-time transcripts of conferences or school, college, and university classes. Respeakers listen to speakers (teachers, MPs, judges, attorneys, TV presenters, etc.) and simultaneously repeat, reformulate and translate the speaker’s saying into a microphone connected to an ASR (Automatic Speech Recognition) software. ASR recognizes a respeaker’s voice and turns the spoken words into writing. For better voice recognition, the words dictated need to be added to the ASR software dictionary beforehand. Thanks to stenomasks respeakers can also work in an environment which is not soundproof.
The Velotype keyboard is a Dutch invention by Marius den Outer and Nicolas Berkelmans, dating back to a mechanical version end of the 1930’s.
Wim Gerbecks and Sander Pasveer modernized the keyboard in 2009.
By pressing several keys simultaneously, complete syllables and words can be made,
instead of typing character by character. The keyboard has a very ergonomical design.
The keyboard can be used in more than 30 different languages and is mainly being used for live subtitling and speech to text.
A velotypist or text interpreter listens to the speaker and normally tries typing at the same speed of the speaker, writing down everything he/she says. But text can also be edited/changed real time in case the speaker is speaking too quickly, speaking unclear or if the setting or audience requires edited output.