Tuesday, February 12, 2013

Weeks 3-4 Vocal tone and r-coloring

During weeks 3 and 4 of the semester I did practice, but I did not do so 8 separate times. Here I record the progress that I did make.

First, I chose to work on several different things related to my vocal tone (going off of the voice quality settings) and ended by analyzing the r-coloring of my speaker.

Days 1 and 2: On these days, I carefully analyzed the vocal tones of my speaker. After comparing my recording from week #2 with the original, I noticed that overal my voice sits at a higher pitch, and there is too much tension in the way I make most of my vowels sounds. I needed to both find a way to relax my jaw so that more of my vowels are rounder and further back, and also carefully monitor when and how my speaker inflects her voice. I started by marking with intonation lines where the voice goes up and down, and highlighting places of extreme rises or falls. I also noted places where the speaker pauses and goes up or down slightly, so that I could get a visual picture of what it is I needed to be saying. I then spent some time practicing this. Below are images of the pages I used for this practice.

            

On days 3-4, I re-recorded with audacity and tried to compare. This was good for comparing length of words and shape of the sound, but not so much for the quality of vowels that I was aiming at.However, close intensive listening to specific sections of the recording and comparing it to my own cause me to notice how very important small changes in stress and tone from word to word in a sentence can alter the understanding of dialect. In those places where I stressed the "wrong" syllables, the vowels are the wrong length and the wave forms different in size. Additionally, I noticed that many of the vowels produced by my Architype have rounder wave forms than mine. In week 5, I plan to look at the two audio files (mine from week 4 and the archetype) and compare them in Praat rather than audacity. Later in the semester, when we discuss rhythm, I also hope to make a chart of the rhythm of the specific words.

Archetype: "It wasn't a recorded message, It was a person"

Me: "It wasn't a recorded message, It was a person"

Lastly, I analyzed the recording for R-coloring. I had noticed that at times the /r/ sounds very american like, and at other times it sounded very British like (r-less). This evaluation caused me to see that "r-full" and "r-less" are not simply an on-off option switch, but rather a continuum. When I listened to my own recording, I recognized myself doing all kinds of funky things with the "r"s to try and sound un-American in my first two recordings, even trilling one of them, but I certainly haven't been making them consistently "Australian". Below are my observations of the patterns.



I practiced this as well. Finally, I made my week 4 recording. Here is the link to it:

Aussie Accent attempt 3

Analysis: Overall I am pleased with my development in this recording. I feel that the vocal quality is better, my r sounds are better, and my vowels are getting closer. I think this is primarily due to my attempts to change the shape of my mouth and look at myself in the mirror to try and imitate my speaker. This has been helping with the vocal shape. My archetype has a "ringing" tone to some of her high dipthongs, and I hit it a couple of times in this recording, although not entirely. There are places where I seem to fall out of charcter, but I've still got some time to work on it.

My plan for next week is to analyze Praat diagrams of my speech compared with my archetype, transcribe key vowel sounds and try to get a feel for the movements in the vowel quadrangle needed to imitate this speaker.

Tuesday, January 29, 2013

Week 2: Listening and Research


 This week, my goal for the project was to try and set the vocal quality settings that I would need to adequately imitate an Australian accent.

According to Celce-Murcia, Brinton and Goodwin (2010), each dialect or language has not only specific vowel and consonant sounds that they use, but also speaker group-wide characteristics of how they position their vocal organs and utilize muscle tension. This has been shown to be important in several research studies, but much of the information about a given language's vocal quality characteristics are "too subjective or unreliable to merit pedagogical application at the present time." (page 32). There was little advice in the text itself on how to go about finding and developing vocal quality settings.  Thus I got the impression that while setting the Australian vocal quality for my own voice would be very important in the development of a convincing accent, I must figure out for myself what those subjective qualities are. So, in this last week (Jan 21-28) I used three strategies to try and discover aspects of a vocal quality setting for my accent.


First, I endeavored to listen to as much Australian English as possible (via movies and YouTube videos). This activity doubled as an effective excuse to have some down time. I spent about 5 hours this week watching  Australian films and TV shows, focusing on an analysis of the female voices and contrastive analysis with my own accent. I justified this also by acknowledging that many methods of teaching pronunciation emphasize listening as a critical first step (Audio-Lingual method, the Direct method, TPR, and even CLT). Celce-Murcia, Brinton and Goodwin (2010) specifically state that listening discrimination is critical, because you must be able to hear the difference in sounds before you can accurately produce them (46). Furthermore, I'm an auditory learner, so I considered this even more critical for my own personal learning style. I hope to continue exposing myself to the Australian dialect in various forms throughout the semester, although not as intensively as I did this week.

Secondly, I spent about two hours this weekend looking up further information/research about Australian dialects and pronunciation. As an academically minded person, I sometimes find that reading research about something helps me to understand it better, which may in some way help me produce it (eventually). Lastly, I listened to and imitated my archetype each day at least once. Next week I will spend less time on the listening and research and more time in deliberate practice to try and set my vocal quality as closely as possible to my archetype.

Here are the most relevant bits of information that I gleaned from my research this week:
          *I first observed that in general, Male actors tended to have stronger, thicker and more identifiably "Australian" accents than women. Additionally, many of the female speaker's vowels were exremely close or identical to some varieties of British English. I later found the following chart from Cox (2006) detailing the differences in vowel quadrangles between male and female teen speakers of Standard Australian English.
This chart shows visually how much lower the low vowels are for female speakers, and how much more "frontwards" and tense the high ones tend to be. I expect that this chart, as well as the several others in Cox's other research articles will help me as I learn to approximate these vowels. 


           *Overall, I observed that this dialect is much closer to British English varieties than American ones, and thus I can see that the vocal quality settings, or where I put my articulators and my muscle tension, may indeed be one of the most obvious distinguishing factors of the Australian accent. The other really obvious distinction is in sentence stress and intonation, suprasegmental features that I will deal with in another post.

            * It seemed to me through listening that the back of the tongue is generally lower and cupped, while at the same time the front of my mouth feels much tighter than when I speak my native dialect. At first I was not sure whether this is a symptom of attempting any new dialect, or if the dialect itself truly requires that kind of vocal posturing. However, the chart above seemed to support these observations. I further found the following website from Macquaire University in Sydney (where Felicity Cox teaches) that contains comparison vowel charts for North American, Australian, New Zealand and RP English varieties. I will spend time studying this chart and making use of it in the future.

http://clas.mq.edu.au/phonetics/phonetics/vowelgraphs/index.html

Finally, here is my second archetype attempt. While working on this second attempt, I focused on keeping my back vowels low with my tongue cupped a bit, tried to copy the way my archetype speakers' mouth looked in the front (very open with front vowels pushed out) and more closely copying her intonation and stress than last time. I also emphasized eliminating final "r" sounds and elongating the stressed diphthongs.


Attempt #2

Although I am still cringing at a lot of the really bad mistakes and poor guesses at individual sounds, I can tell that the overall tone of my voice has improved since last time. In the initial attempt my pitch was much too high and unnatural. I'm closer on the vowels this time than I was before, and I was able to notice some specific consonant sounds that need work as well (especially s's and r's). My goal for next week is to continue to improve my vocal tone, create quality settings, and analyze in audacity the wave forms of the two recordings to compare overall quality.




Tuesday, January 22, 2013

Introduction and Initial Attempt

On this blog, I will post the progress that I make in the next 14 weeks with developing and imitating an Australian accent. This is a project for Pedagogical Phonology class, Spring 2013.
Each week, I will focus on a very specific area of phonology, and will engage in deliberate practice in that area for 30 minutes, four days per week. I will choose my focus area based on what we are currently covering in class, most likely in the following order: 
      • voice quality setting
      • consonants
      • vowels
      • connected speech
      • stress
      • rhythm
      • prominence
      • intonation

Here is the video that I chose as my Archetype. I will focus on the 0:00-1:20 section of the clip:

http://www.youtube.com/watch?v=qAvHr9paSmc

I decided on a dialect of English rather than another language because I felt I would be better able to hear and judge my own progress. Also, I hoped that it would help me develop a better understanding of the quality of vowels and consonants in different varieties of English, which would in turn help me better explain these sounds to future students and help them improve their English pronunciation.

I chose Australian because I think it will be both difficult and interesting. Most of my English dialect exposure has been with American southern, American standard, and (through media) British varieties of English. I wanted to choose something that I was less familiar with so that I would be less likely to "naturally" gravitate toward the correct sounds based on experience and would have to put in significant effort to force myself to do it correctly.

I chose this archetype because it is of a female speaker who is speaking naturally. I did not want to choose a speech (like from a politician), a narrative (like the narrated cartoon Mary and Max), or a male speaker (like Steve Irwin) because I want to experience a natural speech pattern and fluency level that matches my own female voice and is more realistic. When I train students to speak in English I try to help them learn to speak naturally, so I felt it was only fair to apply these methods to myself as
well. Additionally, I liked this video because I find it ironic that I will be imitating a person with an Australian accent discussing her preference for and a peculiarity of said accent.

Here is a link to my initial recording:

http://dl.dropbox.com/u/41456124/Initial%20Attempt%20(1).mov

As my initial recording clearly demonstrates, this attempt of the accent comes off like a bad combination of British English patterns with a few Australian-like diphthongs occasionally thrown in. I've got a lot of work to do. The areas of most significant difference between my recording and the archetype are in vowel length and quality, connected speech (while maintaining these qualities), and rhythm. I have a few things that I need to work on in terms of prominence and intonation, but I expect these last two items will be easier to match than the others. I have chosen a relatively good archetype match for my own vocal quality, so while I do need to work on matching the attitude that is carried in my voice in certain sentences, it is definitely the sounds of the dialect that are going to give me the most difficulty  They are tighter and shorter than British English in some places, but longer and more lax than my American English in other places.

In the next week I will focus on getting closer to an accurate vocal quality setting. I will try to match the tone and register of the speaker and start getting a feel for the ways I need to change the shape of my mouth to make this thing work. I am looking forward to discovering and utilizing online resources for this accent training, and I sincerely hope that by the end of this semester I will be able to produce an 80-second recording that will be only slightly offensive to Australian ears. Or at least sound not so much like a bad British accent.