The Zurich Center for Linguistics (LiZZ - Linguistik Zentrum Zürich), the Language and Space University Research Priority Program (URPP Language and Space), and associated laboratories and institutes at UZH have been jointly granted funds for a significant upgrade of their facilities for research. The Linguistic Reserach Infrastructure (LiRI) project will enable internationally significant research in linguistics, putting Switzerland at the forefront of experimental and Big Data based research.

The LiRI project will improve both laboratory facilities for linguistic research (and related research fields) and technical infrastructure for the analysis of language and speech. The project will harness the latest data science developments to greatly increase data storage, processing and analysis capabilities. It will also enable the collection of naturalistic data in a methodologically rigorous fashion previously only possible in laboratory settings.

The LiRI facilities will be available to all researchers working with language data. When fully operational, it will serve as a national platform for linguistics data storage/processing/science, uniting as many linguistic data resource types as possible in one data center.

LiRI consists of two subunits: the LiRI lab and LIS, the LiRI Information System.

The LiRI lab for experimental research will be equipped with a bundle of data acquisition units, mainly for use in phonetics, psycholinguistics and neurolinguistics. They will be available to researchers and and to external partners for a fee (see regulations for UZH Technology platforms, and the SNSF cost statement for using university infrastructure). Some of the devices are already available. If you are interested to use them, please contact Dr Lei He, the LiRI lab manager, by email: lei.he@uzh.ch.

The main purposes of the LiRI Information System are:

• To channel the large volume of data produced by the LiRI data acquisition units, as well as additional language data and databases from other academic institutions, into standardized, interoperable and open access resources. There will be detailed access rules for partners as well as a minimum standard for the quality of hosted research data, digital assets, and metadata. All tools and data hosted by LiRI will be subject to the FAIR principles of Open Access. 
• To support empirical research from data acquisition to publication.

Data management: LiRI sets up and manages customized virtual machines (VM) servers for the purpose of processing, analysis, long-term storage and backup of linguistic and language-related data. These VM servers are part of the ScienceCloud environment of the University of Zurich and are accessible within the UZH network and worldwide. Additional software can be installed on request and server access can be customized. Data management services are not limited to language scientists working on data acquired by means of the LiRI device and software inventory. Researchers who plan to store, process and analyze their self-collected data are also welcome to make use of the LiRI server infrastructure.

Collaboration and support: LiRI offers advice and technical and methodological assistance to researchers who are developing a project or have questions about how best to collect, process and analyze linguistic data. Support offered can range from occasional guidance to the adoption of the entire data acquisition and processing task.

Examples of available software tools:

• Databases to store and search data collections (e.g. Elastic Search, Maria DB)
• Natural Language Processing such as automatic part-of-speech tagging, syntactic parsing, semantic tagging etc. (based on a UIMA processing pipeline)
• Data crawling/scraping and processing of web sources, batch download of documents, conversion to XML/JSON
• Workflow tool for the management of linguistic annotation
• Providing access to specialized software for semiautomatic data transcription (e.g. spoken to written language)
• Implementation of tools for automatic text analysis and classification (e.g. POS annotation, semantic analysis)
• General machine learning tasks (supervised or unsupervised), Neural Network Learning etc.
• Audio signal processing (e.g. for human speech or animal vocalisations)
• Statistical data analysis

The LiRI project has been in development since 2017 and will continue to evolve and grow in coming years. It will expand its stock of devices and advertise new positions as funding permits. At the same time, it will initiate collaborative research projects and establish long-term partnerships between LiRI and external teams.