This paper describes a complex system developed for processing, indexing and accessing data collected in large audio and audio-visual archives that make an important part of Czech cultural heritage. Recently, the system is being applied to the Czech Radio archive, namely to its oral history segment with more than 200.000 individual recordings covering almost ninety years of broadcasting in the Czech Republic and former Czechoslovakia. The ultimate goals are a) to transcribe a significant portion of the archive -with the support of speech, speaker and language recognition technology, b) index the transcriptions, and c) make the audio and text files fully searchable. So far, the system has processed and indexed over 75.000 spoken documents. Most of them come from the last two decades, but the recent demo collection includes also a series of presidential speeches since 1934. The full coverage of the archive should be available by the end of 2014.
Building a voice-operated system for learning disabled users is a difficult task that requires a considerable amount of time and effort. Due to the wide spectrum of disabilities and their different related phonopathies, most approaches available are targeted to a specific pathology. This may improve their accuracy for some users, but makes them unsuitable for others. In this paper, we present a cross-lingual approach to adapt a general-purpose modular speech recognizer for learning disabled people. The main advantage of this approach is that it allows rapid and cost-effective development by taking the already built speech recognition engine and its modules, and utilizing existing resources for standard speech in different languages for the recognition of the users' atypical voices. Although the recognizers built with the proposed technique obtain lower accuracy rates than those trained for specific pathologies, they can be used by a wide population and developed more rapidly, which makes it possible to design various types of speech-based applications accessible to learning disabled users.
Historical spoken documents represent a unique segment of national cultural heritage. In order to disclose the large Czech Radio audio archive to research community and to public, we have been developing a system whose aim is to transcribe automatically the archive files, index them and make them searchable. The transcription of contemporary (1 or 2 decades old) documents is based on the lexicon and statistical language model (LM) built from a large amount of recent texts available in electronic form. From the older periods (before 1990), however, digital texts do not exist. Therefore, we needed a) to find resources that represent language of those times, b) to convert them from their original form to text, c) to utilize this text for creating epoch specific lexicons and LMs, and eventually, d) to apply them in the developed speech recognition system. In our case, the main resources included: scanned historical newspapers, shorthand notes from the national parliament and subtitles from retro TV programs. When converted into text, they allowed us to built a more appropriate lexicon and to produce a preliminary version of the transcriptions. These were reused for unsupervised retraining of the final LM. In this way, we significantly improved the accuracy of the automatically transcribed radio news broadcast in 1969-1989 era, from initial 83 % to 88 %.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.