Sunday 7 June 2015

How is Phonetics studied?

How is Phonetics studied?

The phonetician Raymond Stetson wrote: Speech is rather a set of movements made audible than a set of sounds produced by movements. 
 
The field of phonetics can be roughly divided into study of the speaker (articulatory), the sound (acoustic), or of the listener (auditory). Each of these divides down further. There's a useful diagram on page 10 of Hewlett & Beck's 'Introduction to the Science of Phonetics'.

Methods in Articulatory Phonetics
 
There are various instruments to help us look at the vocal apparatus during speech.
A real-time or recorded MRI lets us actually watch the vocal tract, and see how it changes during speech. See the video below for an example of this.
 

Live video of movements during speech production (MRI at 20 ms.)



Other methods are a bit more abstract:
  • Ultrasound Tongue Imaging involves sending ultrasound waves through the tongue from various angles, and comparing the time taken to receive the echo. A gap between the tongue and palate will show up in the image as a line.
  • Palatography involves using a colouring agent (such as dye) on a speaker's tongue or the roof of their mouth to identify which part of the mouth is used when producing different sounds. This method is currently used at UCLA (See Where is Phonetics studied?)

When we begin to analyse the frequencies of the sounds we produce, we're getting out of articulatory phonetics and into acoustic phonetics.
Methods in Acoustic Phonetics

Acoustic phonetics is the study of the sound in the air; the way it travels from speaker to listener. We record speech and try to investigate its acoustic characteristics. Even recorded speech is difficult to study though – sound is temporary whether it's recorded or not. We need visual representations of the sound, the simplest of which is the oscillogram (a way of visualising sound waves) and with th
ese graphs we can analyse and compare the frequencies and other properties of speech sounds. (See the example below)

However, speech is very complex, because it consists of many signals, each with their own frequency. Even so, there is still a regularly repeating period. It will repeat with a particular frequency, which we call the fundamental frequency. This is what we recognise as the 'pitch' of an utterance, and depends on the rate of vibration of the vocal cords. There are many complex methods for finding the fundamental frequency of an utterance, but all have some degree of error, especially because the vocal cords don't give a perfectly periodic signal.

Still, if the vocal cords open and close 150 times in a second, the fundamental frequency will be 150Hz. 
  • Changing this frequency is how we make 'He's late?' into a question. But it's not important for giving meaning to the sounds in other ways; that's why an /a/ sound is the same sound whether said by a man or woman. 

Methods in Auditory Phonetics
 


Now we get to the listener. The hearing mechanism is quite well understood, but it's difficult for phoneticians to get a look at it 'in action' as it receives a sound. The instruments are generally too invasive to use, so when they're needed we have to use cadavers.

But it's good to be reminded that hearing's not quite as simple as just using our ears. We can feel vibrations (even if we're deaf) and even vision plays a part.
For example, we find it easier to understand people in person than on the phone, and not being able to see somebody's mouth can be disorienting, especially in a noisy environment, or in a foreign language. 

ECG's and other ways of directly measuring the brain are important - just like in speech production - but a lot of study is still done by exposing subjects to sounds in large quantities and analysing what they say they can hear. By graphing the results of tests like these, researchers can get a picture of where people see one vowel as 'turning into' another. 


Where to next?

References
 
http://www.linguistics.ucla.edu/faciliti/facilities/physiology/static_pal_new/webpal.html
 

Photo Credits:
  • (Oscillogram) http://www.hum.uu.nl/uilots/lab/courseware/phonetics/basics_of_acoustics_1/oscillogram.html
  • (Covered mouth) http://modernreject.com/wp-content/uploads/2011/07/mouth+shut1.jpg

No comments: