I have a long relationship with speech-to-text technology.
In 1998 we had a room in Student Services where students would go to talk to Dragon Dictate. The more they spoke, the less it understood, the more they would laugh, the more it would transcribe their laughing. It was a very popular service as a ‘pick-me-up’.
By 2012 I managed a large collection of contemporary educational oratory -the Oxford Podcasts collection, which includes some fine examples of inspirational rhetoric and clearly communicated ideas. Our interactions with voice recognition software, however, had been frustrating. During the project the team explored various solutions including both automatic translation and human transcription services. We began a project to explore how to best represent the content of our podcasts in text. By focusing on keywords generated by recognition software we were be able to give a searchable interface to users before they listen and represent the amount of relevant content within. Blog post April 2012
7 years later the challenge of making academic audio collections accessible continues to be one which is high in my mind as we roll out lecture recording across the campus at Edinburgh. We’ve been tailoring our Replay roll-out to support the university’s policy for Accessible and Inclusive Learning .
Some people have asked if we are going to have subtitles on our lecture recordings as default. The answer is no, but we are exploring creative ideas on how we could do it.
My experience is that automated speech to text although improving, is not fully there yet. And costs remain prohibitive, so transcripts or subtitles are not automated in the lecture recording system. Specialist language in lectures remain tricky and are often subtitled badly. It is also difficult for the transcription to discern whether the lecturer is quoting, reading, muttering or joking. The kind of ‘performance’ and content some of our colleagues deliver would need a highly nuanced translation. All UK HE struggles with this challenge and colleagues are anxious that their speech is not misrepresented by a poor quality subtitle which might be more confusing for learners. Blog post August 2017
The overarching objective of our new project for 2019 is to establish and evaluate an initial pilot Subtitles for Media service and make recommendations for future sustainability and resourcing.
The initial focus will be on designing and piloting a service which can scale and improve the usability/ accessibility of our front facing media content through the addition of subtitles and transcripts as appropriate. The service design will aim to include all users and will be primarily concerned with publicly available University media content hosted on Media Hopper Create, EdWeb or one of the University’s Virtual Learning Environments.
The project will have three strands:
- Testing the feasibility, viability and cost of a student-led transcription service
A 3-month pilot will allow us to understand what is needed to establish a sustainable programme of work to support our ambitions based on the outcomes of this pilot phase. The students will gain paid work experience and new digital skills. There is already a thriving market in the local region of students who offer proofreading, transcription, audio typing, subtitling and translation services in their spare time and from home. We will work with academic colleagues in the School of Sociology (Dr Karen Gregory) to research the emerging ‘gig economy’ to understand how best to establish an ethical model for piecework in this area.
- Research and Development
The project will strike a balance between evaluating and costing a model for a growing service, and Research and Development to ensure we keep sight of technology trends in this area and understand how they might influence service development over time. We will run a series of events to engage with other organisations and our own technology leaders in this field to ensure we understand and are able to take advantage of technology developments and opportunities for funding or partnerships.
- Improving digital skills and promoting culture change
We aim to move towards a culture where subtitling our media is standard practice at the point of creation, not only because of changing legislation but because it promotes engagement with our media for the benefit of our whole audience, and at the same time promotes digital literacy and digital skills.
In order to achieve all this, the Subtitling for Media Project will:-
- Establish and evaluate an initial pilot service of a student-led subtitling service
- Develop a costed plan for an ongoing service including support and service management
- Make recommendations for future sustainability and resourcing
- Ensure students are trained to deliver a pilot subtitling service
- Create an ethical model for student piecework in this area
- Deliver training and guidance to enable best practice in media creation
- Develop an understanding of current and future technology to support accessibility and ensure our developing service remains in broad alignment
As part of the ISG vision for the University of Edinburgh we aim to support all digital educators in making informed choices about their digital materials. Through this project to establish a new service, staff and students will develop digital skills in creating and using accessible digital materials. Benefits will include supporting staff and students to understand how and why to make learning materials accessible, and development of digital skills in support of wide scale engagement with digital education. The Subtitling for Media Project will establish and evaluate an initial pilot service and make recommendations for future sustainability and resourcing.