Liberated Learning Consortium was founded to research and develop two interrelated applications:
- Using speech recognition technology to automatically transcribe and caption spoken language in educational environments
- Using speech recognition to produce and disseminate accessible multimedia
The Consortium’s ultimate goal is for speech recognition based captioning and transcription of educational media to become a standard for supporting diverse learning needs in learning environments. The Consortium strives to make these approaches widely available to improve access to information for persons with disabilities.
Through collaboration, the Liberated Learning Consortium is dedicated to:
- Collaboratively researching and testing speech recognition in various learning environments
- Evaluating and reporting the impact of speech recognition on learners with and without disabilities, faculty/instructors, and other stakeholders
- Improving the universal efficiency and effectiveness of speech recognition technologies and techniques through applied research and development
- Developing transferable models for successfully implementing speech recognition in a range of learning environments
- Actively raising awareness of the use of speech recognition as an emerging means of improving accessibility
The LLC is hosted by the School for Global Inclusion and Social Development at the University of Massachusetts Boston (UMass Boston).
LLC activities are supported by research agreements between UMass Boston, IBM Research, IBM Human Ability and Accessibility Center, Nuance Communications Inc. and a network of international research partners. UMass Boston hosts the LLC’s infrastructure and is responsible for support systems, communications, and distributing and managing technologies.
The Liberated Learning Consortium emanated from a 1998 world first study spearheaded by the Atlantic Centre of Research, Access, and Support for Students with Disabilities at Saint Mary’s University, Canada. The purpose of this pilot was to establish the technical feasibility of using Speech Recognition (SR) as an alternative to conventional note taking practices for supporting students with disabilities and improving the overall accessibility of traditional university lectures. During this pilot, instructors used commercially available dictation software to digitize the instructor’s speech and instantaneously displayed it as text on a large screen. After class, the SR generated text was edited and made available to students for study purposes.
The pilot revealed that standard dictation SR systems were generally unsuitable for this application. The trials, however, showed that reasonable accuracy could be achieved and more importantly, that students liked the software generated transcripts. This initial experience convinced organizers to pursue a formal study. With funding from the J.W. McConnell Family Foundation, Saint Mary’s University led an applied research effort and produced a baseline understanding of how SR improved accessibility for students, impacted faculty participants, and identified key technological and implementation challenges.
To continue exploration and development of SR based captioning and transcription solutions, a consortium of university, industry, and non-profit partners was established to jointly pursue a mission to make learning more accessible through SR.