We consistently seek improvements in user interfaces so that systems can be accessed in a wide range of situations and by people with disabilities. TRL is well ahead of the industry in this area, particularly in voice technology and accessibility technology for people with visual impairments.
Competency fields
Robust Speech Recognition
Our target is to develop speech recognition technology which recognizesthe spoken word even better than humans can. First, we are studying noise reduction methods, echo cancellation, and target speech enhancement and detection by using multiple microphones. Second, from a language processing point of view, we are studying non-fluency modeling and the acquisition of unknown words to improve the accuracy of transcriptions of spontaneous speech. We are also developing an advanced technology for speech comprehension.
Totally Trainable TTS (T4S)
Our target is to develop a totally trainable Text-to-Speech system. Conventional TTS output is intelligible but unattractive for most users because the characteristics of the original voice have been spoiled. We are using a newly developed stochastic approach from speech recognition, and training the stochastic models by using prosodic features in speech. This method can produce natural human-like synthetic voices, though the applicable domains are still limited.
Speech Analytics
Beyond tools for analyzing written text, technologies for analyzing spoken conversations with clients are required for CRM and compliance checks in various business scenarios. Recognizing natural speech within dialogue is regarded as a difficult problem, but we have made significant progress in the last few years. We are now tackling various recognition problems in real dialogue, such as the conversations in call centers, and we are also developing application technologies such as audio indexing.
IBM Invisible Barcode Technology
With our IBM invisible barcode technology, we can print invisible 2Dbarcodes over printed articles using invisible ink, take pictures of the invisible barcodes using a camera equipped with a special LED (that the invisible ink responds to), use image processing to extract the barcodes, and finally recover the information of the barcodes. Using this technology, we can embed large amounts of information in paper media without worrying about the location or degradation of the appear-ance, and achieve a portalization of the paper media.
Accessibility Technology for Rich Internet Applications
Aiming to provide easy access to IT devices for everyone, including people with disabilities and senior citizens, we have developed support systems and software for information providers, developers, and users with disabilities. Home Page Reader (1997) is a talking browser to assist in non-visual Web access. Our aiBrowser (2007) gives people with visual impairments an opportunity to quickly and easily access dynamic multimedia Web content. The solution aDesigner (2003) supports Web designers in creating accessible and usable pages by visualizing the blind usability and by simulating views for various types of vision impairments. Currently we are studying technologies to evaluate accessibility and to improve usability by targeting various types of rich Internet applications, such as AJAX and Flash.
Accessibility research
Accessibility research aiBrowser
Accessibility research aDesigner
Open Document Format (ODF) Standardization and Tools
The OpenDocument Format (ODF) is a new XML format for next gene-ration office applications. Through our membership in the Accessibility subcommittee of the ODF technical committee, we are contributing to the standardization of the format, especially as regards accessibility improvements. We are also developing various technologies to improve ODF accessibility, such as accessibility evaluation tools and access tools for presentation documents.
