Filled Pause Research Center Home Page "Uh, what do you mean by, uh..., 'filled pauses'?" "Why do, um, people use filled pauses?" "Um, why are you so, uh, interested in filled pauses?" "Just what, um, is your angle in your, uh, dissertation about filled pauses?" "What, um, are your sources of, uh, information about filled pauses?" "Where can I, uh, learn more about filled pauses on the, uh, web?" "Um, who are you?" Send Feedback!

Find out what's new at the Filled Pause Research Center!


I must say that I really dreaded the laborious process of transcribing nearly sixty minutes of speech in detail. In particular, I was concerned about the introduction of error as I marked up the texts with appropriate tags: "Would I remember the tag codes?" "Would I mistakenly stick a tag in the middle of a word rendering both it and the word useless to corpus analysis?", or "Would I accidentally edit within a tag or delete a portion of a tag making it unrecognizable?" Indeed, I had a problem.

However, I was able to develop my own solution to this problem using a Visual Basic application I have nicknamed "Tagger" (see illustration, below).

"Tagger" - main window

Tagger allows me to open, input, and save speech transcriptions. It further allows me to mark-up the transcriptions with my own set of customized tags in an easy two-step process:

  1. Place the cursor in the appropriate place in the text.

  2. Double click the appropriate description in the list at the right and the corresponding tag will be inserted into the text at the cursor location.

Tagger will automatically check that the location is a correct place for a tag. That means, the 'Short Lexicalized Filled Pause' tag <SLFP> can only be placed immediately following a word. If the cursor position does not match this location then an error message is given. On the other hand, when inputting the 'Long Silent Pause' tag <LSP> the program will automatically insert an extra space so that the tag remains separate from surrounding words.

When the cursor enters a tag, the entire tag is selected. This is useful to prevent me from disturbing the internal arrangement of a tag which would render it meaningless for corpus analysis.

Tags are color-coded (red for discourse boundaries, blue for hesitation phenomena, green for discourse intonation boundaries, etc.) so that I can differentiate them from the rest of the text and from each other.

Though it has served my present needs successfully, Tagger still lacks several features to be of real use to others. The tags (my original system) are contained within the code and are unchangeable. In the future I plan to change this so that any user may customize the tags to be used. Furthermore, there remain some minor (but non-critical) bugs to be ironed out. I hope to correct these and code in some additional bells and whistles to give it a somewhat wider appeal and then release a final version before spring. So, if you're interested in what Tagger has to offer, please keep watching this page for a download link.

(Download links coming soon...)

Filled Pause Research Center Home Page

send feedback

This site is maintained by Ralph L. Rose
Last Revised: 99/08/26

Note! This is the original FPRC ca. 1998. It is made available for archival purposes only. Click here to return to the current FPRC.