Research Aim 4: Big Data Sharing and Knowledge Exchange
A one-day Big Speech Data Workshop is to be held at the University of Huddersfield with three objectives in mind:
- To introduce the WY database being collected
- To present different aspects and perspectives on big data for FSS purposes
- To form the first ever international working group for the promotion of big speech data sharing
The one-day Big Speech Data Workshop is proposed to be held in July 2017, and will host a number of invited and contributed presentations. The workshop will culminate in the formation of a working group to promote the sharing of big data. The one-day workshop will be advertised nationally and internationally and is expected to attract academics, and private and public sector workers interested in forensic phonetics, automatic speaker recognition, speech science, sociolinguistics and phonetics. Calls for contributed presentations will be made, and are intended to provide additional perspectives on big data in practice or to introduce an available database.
The one-day workshop will finish with a panel discussion and the formation of an international working group for the promotion of big data sharing.
The key role in the formation of a working group to promote the sharing of big data will be the creation and maintenance of a website. The website will be hosted at the University of Huddersfield and will provide:
- Links to publicly available speech databases with contact information (from across the globe irrespective of language)
- Identification of speech parameters that have previously been analysed in the given database
- Any publications that have come from the aforementioned database
The working group is intended to help in updating the website as new research is published or made available. Members of the working group will be publicised on the website and serve as international ambassadors for the sharing of big data. Members can serve as points of contact for other academics, private and public sector forensic experts if they would like to share data they have collected or if they are trying to obtain a specific type of population data.
