Re: ESPS label files

J Local (lang4@york.ac.uk)
Thu, 23 Apr 1998 11:35:04 +0100 (BST)

Dear Mark - I have no strong feelings about this - I haven't really spent
huge parts of my life labelling in waves - (except that if we're
doing various kinds of overlapping labels it might be advantageous for
searching purposes to have starts and ends). Though I must say that When
I've done labelling and used awk scripts to pull out info it's been a pain
not to have start-end pairs.

John

John Local
Professor of Phonetics and Linguistics
Department of Language and Linguistic Science
University of York
Heslington
YORK YO1 5DD

Tel: UK+ (0)1904 432658
E-mail lang4@york.ac.uk
URL http://www.york.ac.uk/~lang4

On Wed, 22 Apr 1998, Mark Huckvale wrote:

> Dear All
>
> I am in the process of converting the signal and label data into
> ESPS format, so that you can all download it from the the FTP site
> on synth.phon.ucl.ac.uk.
>
> I find that the ESPS label file format, curiously, uses an idiosyncratic
> method of timing labels in label files. I want to know whether you
> use the ESPS standard or something else.
>
> The normal convention for label files is to mark the START and END of
> each region, e.g.
> 0.0 1.1 sil
> 1.1 1.4 s
> 1.4 1.6 t
> However, ESPS, only records a single time - that of the END:
> 1.1 sil
> 1.4 s
> 1.6 t
> Personally, I find this rather confusing. However I am happy to
> go along with it if you (pl) are.
>
> Mark
>
>