Vowel lenition in European Portuguese

(Due _______)

Introduction

For most of you, this will be an unpleasant assignment, brightened perhaps by exposure to some interesting sound changes, but still rendered tedious by the amount of labor that each step requires. You may be able to comfort yourselves by considering how much worse it would have been if you had needed to start by making paper spectrograms, and how much better it will be when all of those tecious steps have been automated.

Preparation

Read these notes on Portuguese Phonology and Phonetics, and this description of the Coral Map Task dataset. Also glance at the other links on this short list of resources, which we'll return to later.

Data

Create a new directory to work in (via mkdir). Change to that directory, and download q2c1m4_0.flac, q2c1m4_0.txt, and q2c1m4_0.TextGrid.

• q2c1m4_0.flac is the audio recording of the Guide's side of one Map Task interaction (of 56);
• q2c1m4_0.txt gives a numbered list of 110 phrases in that recording, with their start and end times followed by an orthographic transcription;
• q2c1m4_0.TextGrid is the same thing in Praat's execrable TextGrid format.

Analysis

I'll send out a message via Piazza assigning each of you to a proportionate fraction of the 110 phrases.

Open the .flac and .TextGrid files in Praat, and add a new Interval Tier named "Phones". For your assigned subset of the phrases, create a phonetic transcription on that tier in Praat, using the ("CODE") symbols in the "Notes on Portuguese phonology and phonetics". Feel free to add new symbols if you encounter phenomena not found in those tables.

Turn in:

1. Your TextGrid file;
2. A table of any additional symbols you use
3. A listing of what happened to each of the (orthographic) vowels in your sample;
4. A summary of the apparent patterns of vowel lention/deletion in your sample.