The Gussenhoven article looked mainly into the physical production of speech sounds, and how the different physical systems interact to create all the possible sounds humans make. Physical speech can be separated into two main parts: the role of the lungs and the larynx and the role of the vocal tract that stems from the larynx. The lungs and the larynx affect speech production by controlling air pressure difference in the vocal tract and larynx. A few examples of this is in the instance of whispering, breathy voice, or creaky voice. The sound from the lungs and larynx can be further modified by the vocal tract. For example, the tongue can be placed on the roof of the mouth to create friction and make a hissing sound. Additionally, I thought it was interesting how inadequate our language represents the multitude of sounds we are able to make. The article introduced a lot of basic concepts that are used to describe more complex ideas in the Kenstowicz article.
The Kenstowicz article mainly looked into the difference between phonology and phonetics as branches of linguistics, and how speakers often falsely judge sounds to be identical when they are actually distinct and vice versa. An example of this is that the coronal stop (t) has eight distinct pronunciations although t is represented by the same letter in English. This can be seen in the plain "t" in stem vs the glottalized "t" in hit. I thought it was really interesting how even though we are generally unaware of these phonetic illusions, we have become adept at obeying the rules the illusions create, and we notice when others break these rules.
Reading both of these articles made me think about how unconscious the act of speaking is and how native speakers "just know" how to speak without being able to analyze how they are doing it. The Kenstowicz article shows that there are phonetic illusions where we hear sound differences that do not exist and perceive similarities where there are none. Despite this, we are able to understand an underlying set of rules and can recognize when they are broken. When they are broken, we perceive someone as speaking weird or having an accent. Even though the word "hat" and "stem" both contain t's, the sound is completely different. "Hat" contains a glottal stop while "stem" contains a plain t, and we know to pronounce them differently without thinking about it. In the Gussenhoven article, we realize there are various systems that interact to create sounds, but when we create a sound, we do not physically think of the organ of speech involved in making the sound, we simply are able to make the sound after hearing it.
These underlying rules that aren't reflected in written language could pose difficulty for a computer that attempts to convert speech to text. For example there are multiple ways to pronounce a "t" including plain t, aspirated t, retroflexed, flapped, nasal flap, glottal stop, or an absent t, and a computer has to recognize all these different sounds a map it to a single letter: t. Although humans are able to learn these nuances without fully being conscious of it, a computer must be programmed to explicitly know these rules and apply them when it makes sense. So for instance it is important to formally understand under what conditions a "t" is a plain t or a glottal stop.
I agree that it's incredibly interesting that so much of language is unconscious -- beyond our ability to recognize and produce different phonemes without thinking about it, linguistic syntax also seems to hold this property (as I'm sure we'll get into later this class). This became especially interesting for me when trying to learn another language: I remember when I was learning Spanish in high school and went on a immersion trip to Spain that I was wildly conscious about every word that was coming out of my mouth, making sure it grammatically and syntactically fit when I was in conversation with a native speaker. I was struck by the realization that the Spanish friends I was making were just as unconscious with their syntax in Spanish as I was about mine in English. It really made the idea of ever being able to be fluent feel so far away -- though I know with enough practice I WOULD start to think about Spanish syntax unconsciously. I'm fascinated by the idea that with enough practice, something as seemingly ephemeral as language can become unconscious.
ReplyDelete