From LinguisticAnnotation
Jump to: navigation, search

This is a review of HIAT-DOS written by Thomas Schmidt for the working group "Transcription and annotation of primary data" at the E-MELD workshop 2006.

Clarification: HIAT (Halbinterpretative Arbeitstranskriptionen - „semi-interpretative working transcriptions“) is the name of a transcription convention. It is not the name of a software tool, but there are several tools that were designed with transcription according to HIAT in mind. HIAT-DOS is one of these tools, the others are syncWriter and the EXMARaLDA Partitur-Editor. The latter offers an import filter for transforming HIAT-DOS data to XML (this will still require manual post-editing, though).

HIAT-DOS is an MS-DOS application, i.e. it will run on Windows, but without making use of the operating system's graphical user interface. It will not run on any other operating system. The main idea of the tool is to support the user in creating transcriptions in musical score notation, i.e. transcriptions where the contributions of different speakers are arranged on separate tiers making possible an intuitive representation of simultaneous or overlapping speech events. Creating and editing such muscial score transcripts in standard text processing software can be tedious because changes, deletions and insertions often make necessary a complete re-arrangement of the whole system wrt line-wrapping etc. HIAT-DOS is meant to help in that task. A transcript that has been created in HIAT-DOS can be exported to RTF or HTML for further editing in a text processing software, printout, web display etc.

HIAT-DOS is, by all standards, an outdated tool. Its interface is completely text-based – it offers no GUI menus, no mouse support etc. Furthermore, it does not offer any support for linking transcripts with the underlying audio or video signal. This alone makes it inferior to virtually every other transcription software featured in this (i.e. the E-MELD) tool room. An even more serious drawback, however, is the fact that HIAT-DOS data fail to meet any of the current best practices for sustainable data handling. The data format of the tool itself is a text-based format which describes the display rather than the content of a transcription. As far as I can tell, the character encoding used in this format is not compatible with any current standard (it may be using an old IBM code page). It is certainly not possible to produce Unicode-compliant data with HIAT-DOS. And it should be very difficult to use the tool with any other alphabet than Standard Latin (with German extensions). The tool's export formats (RTF and HTML) are also 100% display-oriented and do little to make HIAT-DOS data more suitable for archiving. Development of the tool came to a standstill over ten years ago, the source code is not available, and there is hardly any documentation. For all of these reasons, I would strongly discourage any researcher interested in producing exchangeable and archivable data from using HIAT-DOS. To quote Albert Bickford (on IT for MS-DOS): “A great tool when it was released, but now left to rest in peace.”


  • Ehlich, Konrad (1992): HIAT - a Transcription System for Discourse Data. In: Edwards / Lampert (1992), 123-148. In: Edwards, Jane / Lampert, Martin (Hrsg.) (1992): Talking Data - Transcription and Coding in Discourse Research. Hillsdale: Erlbaum.
  • Rehbein, Jochen / Schmidt, Thomas / Meyer, Bernd / Watzke, Franziska / Herkenrath, Annette (2004): Handbuch für das computergestützte Transkribieren nach HIAT. Arbeiten zur Mehrsprachigkeit, Folge B (Nr. 56). Universität Hamburg: Sonderforschungsbereich Mehrsprachigkeit. Available online from At this address, there are also online examples of HIAT transcripts with sound files.
  • Schneider, Wolfgang (2001): Der Transkriptionseditor HIAT-DOS. In: Gesprächsforschung 2, 29-33. Available online from