Some preliminaries…

Before you start, download Phon from the Phon Project website here.  The screencasts below were created using version 1.5.2 which was released on February 29, 2012.

You will also need OpenOffice for doing data analysis which you can download from here.  Phon is also compatible with LibreOffice and, in fact, LibreOffice is probably the better choice in future for a series of boring reasons that are not worth mentioning here.  Note that Microsoft Excel does not display IPA symbols properly when they are exported from Phon.  If you want, you can work in Excel once you’re at the stage of making pretty charts and graphs, but before that you will need the regular expression capability and font-friendliness of Open/LibreOffice.

Please note that your data must be transcribed in Unicode for it to display correctly in Phon.  For most people, this means using the Doulos SIL font which is free to download and install in the appropriate place on your computer.  If your data have been transcribed in SIL IPA93, you have to convert them to Doulos SIL.  SIL International explains how to convert data from SIL IPA93 on the SIL IPA93 Data Conversion page on their website.  It involves downloading their SIL Converter and installing it in MS Word 2003.  This is a straightforward process if you have MS Word 2003.  It is also possible to convert data in plain text files and Standard Format Marker text files.  Otherwise, contact SIL directly regarding conversion (and good luck!).

How to use Phon

Click on the links below to watch screencasts explaining how to use Phon on various topics.

1.  Preparing to use Phon

• Getting your data into Phon (Part 1)
• Getting your data into Phon (Part 2)

2.  Working in Phon

• Entering information about participants
• Linking audio files
• Segmenting audio files

3.  Exporting data from Phon

• Getting all your data out of Phon using merge sessions (which is necessary for word structure analyses).  NB: Merge sessions has been replaced with Export Data as CSV in Phon 1.6.
• Installing the Phone Clusters script (which is necessary for any kind of consonant analysis)  Note: If you are using a Mac and cannot find the Library, it is likely just hidden.  To unhide your Library, go to your Applications folder, then to your Utilities folder.  In the Utilities folder, double click on Terminal.  A window will open up with a black background.  Copy and paste the following command into the Terminal window: chflags nohidden ~/Library then press Enter and your Library should magically reappear.
• Using the Phone Clusters script for singleton consonant analysis by word position
• Using the Phone Clusters script for consonant cluster analysis (including how to find and use the template)
• Using the Phone Clusters script for feature analysis

How to analyze your data using spreadsheets

Whole word match and word shape analyses

• Data preparation and introduction to the template
  
• Conducting a whole word match analysis using the template
• Tabulating whole word match results
• Conducting a word shape analysis using the template
• Tabulating word shape match results

Word length and stress analysis

• Conducting a word length and stress analysis using the template
• Tabulating word length and stress results

Singleton consonant by word position analyses

• Conducting a singleton consonant by word position analysis using the template
• Tabulating singleton consonant by word position results

Consonant cluster analysis

• Tabulating consonant cluster results by group
• Tabulating consonant cluster results by subject

Feature analysis

There are a series of three templates (one for each word position) which automate, to a large extent, the calculation of the number of subjects in a group who show 90% accuracy or more in their production of a given phoneme.  It also calculates the percentage match by phoneme according to left and centre-right prominence.  The formulas used to do these calculations are complex.  Too complex, it turns out, to be converted from OpenOffice Spreadsheet to Excel without errors.  This is a problem because, while Excel spreadsheets can be uploaded to the blog, OpenOffice spreadsheets cannot.  If you are interested in trying out this template, please contact someone in the lab for a copy.  If you are curious about how they work and would like to see how to use them, watch the Tabulating consonant mastery by word position screencast.

Resources

The talented and helpful developers of Phon wrote a script for the SASS phonology lab to facilitate analysis of both singleton and consonant clusters according to word position and stress pattern.  Download the script and accompanying documentation in this zip file.  Once you have downloaded it, please refer to Installing the Phone Clusters script screencast for installation on Macs.  For installation on Windows, see the Phon script library page.

Several templates in the form of Excel spreadsheets that are compatible with Open/LibreOffice have been made to expediate data analysis.  The columns are already set up and formulas have been included so that calculations, such as percentage matches, can be done quickly and easily.  The templates are provided below and are organized by analysis type.  Please refer to the appropriate screencast for the various spreadsheet tips and tricks you will need to be able to make efficient use of the template.

• Whole Word and Word Shape Match Analysis Template
• Word Length and Stress Analysis Template
• Singleton Consonant by Word Position Analysis Template
• Consonant Cluster Analysis Template

Troubleshooting

• Dealing with Affricates in the Word Shape Match Analysis