Vienna-Oxford International Corpus of English (VOICE)

At a time of ever-increasing globalisation and mobility, developing an understanding of intercultural communication is a highly valued asset. The fact that `English` features like no other language in global communication is commonplace - what is usually not made explicit, however, is that this `English` means `English as a lingua franca`, ELF for short. All over the globe, ELF is used on a daily basis by millions of `non-native` speakers from different linguacultural backgrounds. The implication of this state of affairs is that ELF deserves to be conceptualized as a language use in its own right. It follows that linguistic descriptions of systematic salient features of ELF are needed to foster an understanding of the nature of ELF interactions. A Translational Research Project running from 2005 to 2008 (L116-G03) therefore took the first significant steps to compile a computer-readable 1 million-word corpus of spoken interactions in English as a lingua franca. The focus of this corpus is on unscripted, largely face-to-face communication among competent speakers from a wide range of first-language backgrounds. The speech events being captured come from professional, educational and leisure domains and include speech event types such as working group discussions, meetings, interviews and informal conversations. This corpus has provided the first basis for empirical investigation and description of these interactions by the research team that compiled it. However, it is important that this rich resource should be made maximally useful to the scientific community for analyses to be carried out by all researchers that wish to do so, and for applications to be developed that can be geared to specific local and global requirements. The first release of this corpus, VOICE1, will be made available in 2008. While the data will be richly marked up with various kinds of contextual information, in order to facilitate access to the rich potential of this resource, a more comprehensive and linguistically annotated corpus needs to be developed. This is the principal aim of the proposed follow-up project and will involve part-of-speech tagging, lemmatization and other kinds of annotation, such as certain pragmatic phenomena, the development and implementation of which will be a serious but fascinating challenge due to the new linguistic terrain that ELF represents. This second release of the corpus, VOICE2, will also offer the facility of accessing selected sound files so that users can listen to a variety of ELF speakers via the web and analyse, e.g., phonological aspects of ELF. VOICE2 will also be bigger (between 1.5 and 2 million words) and cover a greater variety of geographical areas/first language backgrounds and more domains of use. The need for such a comprehensive and accessible resource for research on ELF and the potential of the expected applications would seem to be self-evident. The global and globalising force of English is a burning issue in today`s society, and so ensuring open and fully exploitable access to the corpus to foster applications based on a sound research basis is of paramount importance.

The main objective of this project was to respond to the demand for a comprehensive and accessible open access resource for research on English as a Lingua Franca (ELF) and to make possible applications based on it. At a time of ever-increasing globalisation and mobility, developing an understanding of intercultural communication is a highly valued asset. The fact that English features like no other language in global communication is commonplace what is usually not made explicit, however, is that this `English` means `English as a lingua franca`. All over the globe, ELF is used on a daily basis by millions of `non-native` speakers from different linguacultural backgrounds. The implication of this state of affairs is that ELF needs to be conceptualized as a language use in its own right. It follows that linguistic descriptions of salient features of ELF interactions are required to provide for an understanding of the nature of ELF communication. The result of the preceding project L116-G03 (2005 2008) was a computer-readable 1 million-word corpus of spoken interactions in English as a lingua franca. The focus of this corpus is on unscripted, largely face-to-face communication among speakers from a wide range of first-language backgrounds. The speech events captured come from professional, educational and leisure domains and include speech event types such as working group discussions, meetings, interviews and informal conversations. This corpus provided the first basis for empirical investigation and description of these interactions. It is of crucial importance that this rich resource should be made maximally useful to the scientific community for analyses to be carried out by all researchers that wish to do so, and for applications to be developed that can be geared to specific local and global requirements. This was the main objective of the follow-up project L 448-G03 (2008-2013). The VOICE corpus was first released in 2009, with a user-friendly online search interface and freely accessible via the internet. This version already contained rich mark-up with various kinds of contextual information and full documentation. In order to further facilitate access to the rich potential of this resource, a more comprehensive and linguistically annotated corpus was developed. In a second step, access was enabled in 2010 to selected anonymized sound files so that users can listen to a variety of ELF speakers via the web and analyse, e.g., phonological aspects of ELF. Probably the most significant developments of VOICE for advanced corpus users followed in the next phase. These involved the release, in 2011, of VOICE for download in XML format, followed in 2013 by the release of VOICE POS, that is, the corpus enriched by part-of-speech tagging, lemmatization and other kinds of annotation, such as certain pragmatic phenomena (e.g. discourse markers). The development and implementation of these improvements proved a considerable but fascinating challenge due to the new linguistic terrain that ELF represents, and new ways of thinking that are needed to understand it. In order to cover a greater variety of geographical areas and first language backgrounds, cooperation was realized with various Asian universities, resulting in the compilation of the Asian Corpus of English (ACE). ACE has the same architecture as VOICE, thus making it possible to investigate twice the amount of data in parallel and for comparisons to be made. The global and globalising force of English is a burning issue in today`s society, and so ensuring open and fully exploitable access to the corpus to foster applications based on a sound research basis is of paramount importance.