About the BAS
ContactBayerisches Archiv für Sprachsignale
c/o Institut für Phonetik, Universität München
Schellingstr. 3 / II
Telefon: +49 (0) 89 / 2180 - 2758
Fax: +49 (0) 89 / 2180 - 5790
- Foundation: 01.01.01.1995
- Financing: fundings by the Bavarian State, the University of Munich and cooperations
The Bavarian Archive for Speech Signals (BAS) is a public institution hosted by the University of Munich founded with the aim of making speech resources of contemporary spoken German available to research and speech technology communities via a maximally comprehensive digital speech-signal database. Speech material will be structured in a manner allowing flexible and precise access, with rich annotations, metadata and linguistic-phonetic evaluation forming an integral part of it.
The last decades have seen an abrupt increase in the demand for large speech-signal data collections, both on the part of academic investigators carrying out basic research as well as on the part of engineers from industry working in the new integrated field of speech and information technology. There are many reasons for this. Primarily, however, the growing increase in demand must be attributed to the breakneck pace of hardware and software development in speech signal processing. The increasing number of techniques for acoustic-phonetic signal processing, and the increasing amount of speech data that can be efficiently handled and processed together generate an accompanying demand not only for linguistically interesting text material (which of course emerges automatically from the modern printing industry) but also for reliably acquired and phonetically evaluated spoken language material. A number of national and international initiatives (such as BDSON, PHONDAT, LDC, SPEX, COCOSDA, METANET, CLARIN) have already resulted in the collection and distribution of large speech corpora. However, they exhibit a variety of formats, corresponding to the variety in the aims pursued. For German, a central institution was clearly lacking that could carry out such tasks within a long-term perspective. BAS will be responsible in Germany for these tasks for distributable resources of spoken German, collecting, maintaining and making them available in standardized form.
In addition, BAS will develop its own tools for automatic recording, linguistic processing, labelling and segmentation, making the results available via public domain software packages and/or web services.
BAS was entrusted by the Bundesministerium für Bildung, Wissenschaft, Forschung und Technologie (BMBF) with the task of maintaining both existing and future databases set up within funded projects by the BMBF, and of exporting them (after any restrictions on availability have expired) within the EU as well as to the Linguistic Data Consortium (LDC). Imported databases are to be converted by BAS to a standardized form (CMDI), enabling them to be expoited in all BMBF- funded speech projects for a fraction of the cost and effort usually incurred.
BAS Repository of German Speech ResourcesThe BAS Repository is organized according to the FAIR guidelines for scientific data: Findability, Accessibility, Interoperability, Reusability:
- Findability (i): Speech resources archived at BAS will be visible in science indices and meta search engines. The metadata (MD) of archived resources follow common MD standards such as DC, OLAC and CMDI, and are exported via a public OAI-PMH interface to harvesting science indices, e.g. Web of Science Data Citation Index, Virtual Language Observatory (https://vlo.clarin.eu/), or Open Language Archives Community (http://www.language-archives.org/). BAS metadata are always public and must not contain any personal data.
- Findability (ii): Speech resources archived at BAS can be referenced by a permanent URL called Persistent Identifier (PID). Each corpus (and each recording session within) are assigned to PIDs of the handle system (https://www.handle.net/) that are maintained by the BAS. PIDs allow citations to data sets that will always point to the corresponding repository page or MD record even if the archive has moved to another web address.
- Accessibility (i): Speech resources archived at BAS are protected against unauthorized access. The owner of the resource decides according to three basic CLARIN lincense models who can access her data: ''PUB'' means the resource is available for everyone, ''ACA'' licensed resources are accessible only for academics that can authentificate themselves via the international AAI Shiboleth system, and ''RES'' resources are restricted to a list of users maintained by the owner.
- Accessibility (ii): Speech resources archived at BAS are protected against data loss. The BAS Repository is organized according to Open Archival Information System (OAIS) principles (biannual Core Trust Seal, CLARIN Center B evaluations); daily backups via encrypted channels are run to the Leipniz Rechenzentrum Munich and to Rechenzentrum Jülich. Only authorized IT personell at BAS have direct access to archived data.
- Accessibility (iii): Speech resources archived at BAS can be accessed any time and without delay by the owner of the data. Data are stored on 'LiveSystem' (spinning discs), not on LTO tapes.
- Interoperability: Speech resources archived at BAS are checked periodically for integrity and whether their data formats are still supported by state-of-the-art software. If necessary, archived media streams are transcoded to a new loss-less format.
- Reusability: Speech resources archived at BAS can be a valuable source for other disciplines; by means of the BAS repository they are visible in that regard, especially for speech technology applications.
The Leibniz Rechenzentrum München (LRZ) -- which is connected to the site via fiber optic data link -- provides the Archive with mass storage and network support within the framework of the TERABACK project.
Since June 20th, 2013, the BAS is a licensed CLARIN Center of type B; the CLARIN centers throughout Europe share the same principles and standards as the BAS to maximize inter-operability in the future.
The BAS is keen to cooperate with all institutions in the German speaking area interested in contributing to the common goal. Most of the projects will be financed by interested partners in industry, by public grants or by European projects.
The BAS produces speech resources either by public funding or industrial cooperations. Speech resources funded exclusively by public money are available without restrictions immediately after the release for everybody. Industrial partners that have significantly contributed to the production of the resource are granted a period of one year after the release to exploit the data exclusively. After that period the resource is distributed via the BAS either unrestricted or under license.
Christoph Draxler studied Computer Science at the Technical University of Munich, Germany. He earned his PhD in 1991 from the University of Zurich, Switzerland in the field of databases. Since 1991 he has been working at the Institut für Phonetik und Sprachliche Kommunikation mainly within the PhonDat and VERBMOBIL projects. His main interests include logical programming, databases and multi-media applications.
Florian Schiel received his Dipl.-Ing. and Dr.-Ing. degrees from the Technical University in Munich in 1990 and 1993 respectively, both in electrical engineering. Since 1993 he has been with the Institute of Phonetics, University of Munich, participating various large BMB+F projects. His main interests are: speaker charcteristics, German phonetics, computational phonology, automatic analysis of very large speech corpora.
BAS Long-term and disaster plans
The BAS aims to be a permanent institution and to offer its archived data and services for an unlimited time. To address long-term data preservation as well as several disaster plans please refer to the public Long-term Preservation Plan.