Enhancing the usability of real-time speech recognition captioning through personalised displays and real-time multiple speaker editing and annotation
read more
Citations
Using Speech Recognition for Real-Time Captioning and Lecture Transcription in the Classroom
Automated generation of ‘good enough’ transcripts as a first step to transcription of audio-recorded data:
Synchronised Annotation of Multimedia
E-Scribe: ubiquitous real-time speech transcription for the hearing-impaired
A Speech-To-Text System’s Acceptance Evaluation: Would Deaf Individuals Adopt This Technology in Their Lives?
References
Speech recognition in university classrooms: liberated learning project
Speech-Based Real-Time Subtitling Services
Creating accessible educational multimedia through editing automatic speech recognition captioning in real time
An exploration of the potential of Automatic Speech Recognition to assist and enable receptive communication in higher education
Using Automatic Speech Recognition to Assist Communication and Learning
Related Papers (5)
Frequently Asked Questions (10)
Q2. How many errors per minute would be required to correct?
Somebody talking at 150 words per minute with a 22% error rate produces an average of 33 errors per minute and if correction of only 20% of these errors were ‘critical’ to understanding then the editor would have to correct on average only about 7 errors per minute.
Q3. What is the main purpose of this paper?
Real time captioning (i.e. creating a live verbatim transcript of what is being spoken) using phonetic keyboards can provide a live transcription for deaf people and can cope accurately (e.g. >98%) with people talking at up to 240 words per minute but is often not available because of the cost and shortage of highly skilled and trained stenographers [1] [2].
Q4. What is the purpose of this paper?
To improve accuracy of verbatim captions created directly from the voice of the original speaker the application RealTimeEdit (RTE) was developed to enable corrections to ASR captions to be made in real-time [10].
Q5. What is the effect of a live transcription of speech?
Detailed feedback from students with a wide range of physical, sensory and cognitive disabilities and interviews with lecturers [6] showed that both students and teachers felt this approach improved teaching and learning as long as the text was reasonably accurate (e.g. >85%).
Q6. What is the way to provide a live transcription of speech?
Standard speech recognition (SR) software (e.g. Dragon, ViaVoice [3]) was found to be unsuitable for live transcription of speech as without the dictation of punctuation it produced a continuous unbroken stream of text that was very difficult to read and comprehend.
Q7. How many corrections per minute can be achieved by a user of a text editor?
The unedited outputs of ViaScribe can be merged and then edited as shownin figure 8 or if preferred, each instance of ViaScribe can have its output edited using RTE and then all the edited RTE outputs can be merged;Trials of the system in a variety of settings are being conducted to investigate in practice the effect of error rates, number of speakers, editing operator skill requirements etc.
Q8. How many corrections per minute can be achieved by a single user?
The combination of ViaScribe, ViaScribe server, PDC, RTE, and RTM enables a very flexible approach to be adopted that can provide solutions to many requirements.
Q9. What is the main purpose of the paper?
Text transcriptions of the spoken word can benefit deaf people and also anyone who needs to review what has been said (e.g. at lectures, presentations, meetings etc.)
Q10. What is the purpose of the paper?
While projecting the text onto a large screen in the classroom has been used successfully in LL classrooms it is clear that in many situations an individual personalised and customisable display (e.g. font size, formatting, colour etc.) would be preferable or essential and so a personalised server and client was developed to enable users to customise their displays on their own networked computer [7].