SIL International Home

Phonology Assistant

Current version: 3.0.1, released 2008

This new version of Phonology Assistant provides support for Unicode. In addition, PA is designed to support Speech  Analyzer, Toolbox, and FieldWorks Language Explorer users with a phonology tool that works interactively with the data stored in those applications.

Phonology Assistant creates consonant and vowel inventory charts and assists the user in searching through the data corpus to do phonetics and phonology.

Features

Source data from Toolbox, FieldWorks, or Speech Analyzer

New: Use data transcribed with the Unicode IPA character set from:

  • Speech Analyzer files with transcriptions.
  • Toolbox databases (interlinear and non-interlinear).
  • FieldWorks databases. FieldWorks supports capturing phonetic data in the lexeme form or as a pronunciation. Two exports are provided for use with Phonology Assistant.
  • Any application that can export data to XML and apply an XSL transform to conform to PAXML.
  • Any combination of the above.

Note: Phonology Assistant does not have a data corpus editor. It is assumed that users will use one of the above programs.

Phonology Assistant can jump to the source record in any of the above SIL programs so that the user can view and edit the source record. When changes are made and saved in the source data, PA automatically updates charts and word lists to reflect the changes.

Click thumbnail for full-sized image
Click thumbnail image for full-sized image

Uncertain data

New: Unsure of how to transcribe a word? Was it [I] or [i]? Transcribe both of the uncertain alternatives. Views in the corpus, charts and search results reflect each of the uncertain transcriptions. Triangles in the corner of a cell help you know when you have uncertain transcriptions.

  • PA indicates to the user when source data has unusual characteristics the user should be aware of, in this case, an uncertain transcription.
    Click thumbnail for full-sized image
    Click thumbnail image for full-sized image
  • Uncertainties can optionally be accounted for in charts and word lists.

Ambiguous sequences

It is important to call this "ambiguous sequences" and not "ambiguous data." Here are some screenshots and an explanation.

  • Unsure as to whether or not a consonant cluster should be treated as one or two phones? PA allows you to experiment with different interpretations of ambiguous sequences. Below, the sequence kp is treated as one Consonant.
    Click thumbnail for full-sized image
    Click thumbnail image for full-sized image
    This treatment is reflected in the search results as shown in the CV Pattern field and consonant chart below. If this treatment turns out to not make analytical sense we can easily uncheck the "Treat as one unit?" checkbox to see the effects of the other interpretation.
    Click thumbnail for full-sized image
    Click thumbnail image for full-sized image

Experimental transcriptions

New: Experiment with potentially ambiguous items or sequences without changing your transcription.

Example: if your source transcription contains [ts], you may want to temporarily treat it as [t͡s ] in your charts and word lists to test if it is best treated as one or two consonants.

Click thumbnail for full-sized image
Click thumbnail image for full-sized image

Charting

  • View a chart of all consonant or vowel phones found in a given phonetic data corpus.
  • View a histogram to see frequencies of phones in the data corpus.
    Click thumbnail for full-sized image
    Click thumbnail image for full-sized image
  • Add, delete or modify columns and rows of phonetic charts and place phones where you want them to appear in the chart.
  • See all phonetic data items containing a particular phonetic segment and sort preceding and following contexts from left to right or right to left.

Phonetic searching

  • View multiple search result word lists at once for easy comparison of phones in their environments.
    Click thumbnail for full-sized image
    Click thumbnail image for full-sized image
  • Create simple or complex phonological search patterns.
    • Search patterns can be typed in directly (or pasted from an email from your linguistic consultant!)
    • Point and click using the Search Pattern builder
      Click thumbnail for full-sized image
      Click thumbnail image for full-sized image
    • Create AND / OR search patterns with a variety of wildcard characters
      Click thumbnail for full-sized image
      Click thumbnail image for full-sized image
    • Diacritics, stress, syllable breaks, etc. can optionally be ignored in searches.
      Click thumbnail for full-sized image
      Click thumbnail image for full-sized image
  • Use predefined articulatory or binary features in searches.
  • Create user-defined classes for use in searches.
  • Sort by search item, preceding or following environments.
    Click thumbnail for full-sized image
    Click thumbnail image for full-sized image
  • Group data by any column including CV pattern, speaker, preceding or following phonetic environment.
  • Easily find minimal pairs.

XY charts

  • XY charts display rows of search items and columns of search environments.
  • Each cell displays the number of occurrences of an item in an environment.
  • Chart definitions can be saved for easy re-use.
    Click thumbnail for full-sized image
    Click thumbnail image for full-sized image

Export

  • Export charts and search results as HTML or RTF for incorporation into web pages or pasting into word processors.

Sound

  • New: Listen to source sound files from Speech Analyzer transcriptions, or if source is indicated in Toolbox or FieldWorks database.
  • Use Speech Analyzer playback capabilities to increase or decrease playback from within Phonology Assistant.

Note: If you have transcribed data that cannot be converted to Unicode, Speech Tools 2.2 contains non-Unicode versions of Phonology Assistant and Speech Analyzer that are still available at Speech Tools downloads.

Phonology Assistant Support