MD recorder

[Steven Bird]

Karen Nakamura <karen.nakamura at yale.edu> writes:

> The big minus is the lack of a professional MD transcribing machine (foot
> switches, variable speed, etc.). As MD catches on, I think we'll start to
> see them.

A better solution for some applications is to upload an entire
recording and transcribe it using suitable software.  A particularly
good tool is "Transcriber" [www.ldc.upenn.edu/mirror/Transcriber/],
which is designed to handle extended (e.g. 2 hour) audio files and
understands most speech file formats.

A major benefit of this approach is that you get a *time-aligned*
transcription, which is easy to verify and correct.  And since the
transcript does not have to reside with the audio, large audio files
can be kept on CD-ROM.

This kind of transcription is a special case of "signal annotation",
where any kind of audio, video or physiological signal is annotated
(or coded, or marked-up) with descriptive or analytical notations.

For more information about initiatives in this area, please see the
Linguistic Annotation page [www.ldc.upenn.edu/annotation/].

--
Steven.Bird at ldc.upenn.edu  http://www.ldc.upenn.edu/sb
Assoc Director, LDC; Adj Assoc Prof, CIS & Linguistics
Linguistic Data Consortium, University of Pennsylvania
3615 Market St, Suite 200, Philadelphia, PA 19104-2608