Captioning Multiple Speakers Using Speech Recognition to Assist Disabled People

Abstract

Meetings and seminars involving many people speaking can be some of the hardest situations for deaf people to be able to follow what is being said and also for people with physical, visual or cognitive disabilities to take notes or remember key points. People may also be absent during important interactions or they may arrive late or leave early. Real time captioning using phonetic keyboards can provide an accurate live as well as archived transcription of what has been said but is often not available because of the cost and shortage of highly skilled and trained stenographers. This paper describes the development of applications that use speech recognition to provide automatic real time text transcriptions in situations when there can be many people speaking.

Development of a face-to-face meeting capture and indexing process

“…Whereas, the automatic verbatim captions created directly from the voice of a speaker using automatic machine technology has a much higher word error rate, sometimes in excess of 50%. Editing software therefore has been developed so that human intervention can edit the output caption errors and add in speaker name identification details [15]. Another approach to speech transcription has been presented.…”

Contributions of an Augmented Reality Musical System for the Stimulation of Motor Skills in Music Therapy Sessions

“…Systems for speech recognition (Wald, 2008), (Hua;Lieh-ng, 2010) and eye tracking (Jacob, 1991), (Chen;Pu, 2010) are used to provide access to reading, writing and communication. Brain-computer interfaces (Wolpaw et al, 2002) help people with severe physical disabilities to communicate and move around.…”