Universidade de Vigo uses Dragon speech recognition to innovate training
The GALMA research group at Universidade de Vigo is using Dragon Professional Group speech recognition supplied by Exonera for a course on interlingual respeaking; the course trains students to provide interlingual subtitles for live TV programmes and live events, making them accessible for a wide audience (hearing, deaf and hard of hearing, the elderly, children, foreigners, migrants, refugees).
Hayley Dawson, research group member and trainer at GALMA, the Galician Observatory for Media Accessibility where the course is delivered explains…
“Respeaking is a method of live subtitling in which a respeaker listens to the sound of a live programme or event and respeaks it (essentially repeats it), including punctuation marks, to speech recognition software. The software turns the speech into text displayed on screen.
Dragon speech recognition software has been an essential tool in this course, as trainees have used it for respeaking exercises. To complete the interlingual respeaking exercises, trainees listen to an audiovisual text in English and respeak it (translate it live) into Spanish. To simulate the effect of live subtitles, trainees place a reduced sized window of DragonPad underneath the videos so their respoken output scrolls below the video”.
The main aim of the course is to address the growing demand for professional interlingual respeakers, who are needed to make live content accessible.
The course is comprised of three modules:
- simultaneous interpreting English-Spanish
- intralingual respeaking Spanish-Spanish
- interlingual respeaking English-Spanish
Hayley concludes: “Listening and typing live can be too slow. Using Dragon is a fast method of producing live subtitles to increase accessibility to live content”.
For more information http://galmaobservatory.webs.uvigo.es/respeaking-course/
Or enjoy this video...