It's probably not a good idea to choose a thesis idea as complex as analyzing a language I didn't grow up speaking or hearing. My native language is Thai, yet as a member of the global design community, I also must interact with English, an international language.

I love listening to stories, and I am interested in analyzing the way people tell them. I asked some of my classmates to tell me stories about their childhoods. Faint memories came back as rich stories.

I used a video camera to record the stories as people told them. I transcribed this spoken language into text. Then I created various systems to analyze the elements of storytelling. Finally, I designed posters to represent my analysis. I named my project “Seeing Um Telling” to represent my effort to translate talking into typography.

The mysterious process of speech fascinates me. How do we transform our thoughts into spoken language? During speech, we think, talk, hesitate, pause, and say “um,” or “uh,” to buy time. Talking is an activity that relies on thinking in the moment. It’s part of our communication style and individual verbal habits. All these behaviors affect how each person narrates.

Supisa Wattanasansanee
MICA Graphic Design MFA, 2011



I was born and raised in Thailand however, Thai culture does not limit my work as a traditional label, rather, it provides my work with a wide range of cultural vitality. Everything has its own worth and sets its own trend and future direction. I believe that good design should convey a message on its own while still being attractive to all.

I am a freelance graphic designer in Baltimore. I am currently graduating with an MFA in Graphic Design at the Maryland Institite College of Art (MICA).
Seeing um Telling
MFA Graphic Design Essay
Maryland Institute College of Art
Spring 2011

It's probably not a good idea to choose a thesis idea as complex as analyzing a language I didn't grow up speaking or hearing. My native language is Thai, yet as a member of the global design community, I also must interact with English, an international language.

I love listening to stories, and I am interested in analyzing the way people tell them. I asked some of my classmates to tell me stories about their childhoods. Faint memories came back as rich stories.

I used a video camera to record the stories as people told them. I transcribed this spoken language into text. Then I created various systems to analyze the elements of storytelling. Finally, I designed posters to represent my analysis. I named my project “Seeing Um Telling” to represent my effort to translate talking into typography.

The mysterious process of speech fascinates me. How do we transform our thoughts into spoken language? During speech, we think, talk, hesitate, pause, and say “um,” or “uh,” to buy time. Talking is an activity that relies on thinking in the moment. It’s part of our communication style and individual verbal habits. All these behaviors affect how each person narrates.

After collecting ten stories, I choose three that I thought had potential for analysis. My friends’ stories and the way they narrated them portray their personalities. I started to create systems around the analysis. Starting with three stories gave me a chance to see how each system could apply to other stories. The process also allows comparisons among the stories.

Chris’s story is long, confessional and had the most emotional contrast, while Beth and Ann’s stories are more direct, clear and focused. I chose to focus on Chris’s story, analyzing in twenty graphic ways.

First Design Strategy

This system represents speech patterns. I broke down the story into sections that indicated 5-second intervals. I divided the words into a grid with equal spacing. The number of words in each line forms a bar graph that shows the speed of speech. It showed how fast or how slow Chris talks throughout each session. I also coded each word by color based on the part of speech.

This pattern symbolizes the frequency of speed, tone, rhythm, and fillers. The pattern allows viewers to see the number of times Chris mentions one person and what kind of description he gives that person. Also, the intensities of each color show in what part of the story he uses more fillers and the frequency of negative and positive words.

In the beginning of the story, Chris says, “Ross is the best thing I had.” A reader might interpret the tone of Chris’s statement as positive, but upon close inspection, the reader will see that Chris intended sarcasm. The readers can see the negative colors around one spot of positive colors. The outcome expresses emotion not only through choice of words but also through rhythm and pace during different parts of the story. The pattern allows the viewers to see the mood and tone in each part without reading through the whole story.

After exploring the parts of speech in objective ways, I became more interested in analyzing subjective aspects of the stories. I categorized emotions throughout the story. I was interested in how a computer algorithm would judge the emotions of the story. I filtered the story with a natural language processor, and it pointed out the positive and negative words.

It’s fascinating to compare the nuances of human judgment to the computer filters. For instance, the computer program defined the word pretty as differently as a positive adjective. Yet Chris was using the word as a noncommittal adverb. Because of this, the program categorized the word differently than I did. The program also picked out the word super as positive, but in this context, it is being used as an adjective in the phrase “super freaked out.” By contrast, the program fails to identify the phrases anti-social and no friend as negative while identifying friend and adventure as a positive word. My design represents human judgment with a thick line and the program’s judgment with thin.

The next poster applies an emotional filter to each part of the story. There are four main emotions: negative, contrary, neutral, and positive emotion. I used my own judgment to begin and later asked the story’s teller to verify. I used color to differentiate between certain stages of emotions. Uncovered words indicated fillers such as um, uh, and, like, which contrasted with the emotion fillers. In general, the fillers mostly occurred in the beginning and middle parts of the story, rather than at the end. More important, the fillers are often used when the speaker is switching emotions. I made the fillers more evident in the preceding poster by blocking out other words and emphasizing the fillers.

Second Design Strategy

A key part of my thesis exhibition in the graphic design in Master of Fine Art at MICA was the creation of motion graphics. The video served as an introduction to the posters that follow with the details. The type appears on the screen one word at a time, synchronized with the voice of the tellers. These videos allowed visitors to both see and hear the stories simultaneously without any distracting of images.

The words appear and disappear, symbolizing the spoken language’s transience as it passes from the speaker to the

audience. In contrast with subtitles in a film or closed captions on TV, these motion graphics follow the spoken stories in real time. The screen appears blank, implying that the teller paused. Words that appear for a very short moment mean when the speaker talks very fast.

These motion pieces were simply designed. They presented small black words on a white screen. This simplicity contrasts with the complexity of the color-codes posters. The bar graphs under the screens supplemented the motion pieces. They immediately showed information about different kinds of words that the tellers used in their stories. There were no numbers for each category but only the lengths for comparison.

Third Design Strategy

The last piece of my poster design compared all three stories over time. Each line length indicates the number of words spoken in five seconds. It compares each 5 seconds of each story together. Chris Clark spoke both the fastest with 26 words in 5 seconds and the slowest with 2 words in 5 seconds. He spoke fast during the climax of the story, where he took action after having victimized from the beginning. The slowest part of his story was when he lost his train of thought in the story’s transition section. The poster shows how long, fast, and slow he talked, and how long he paused. Chris’s variable data makes him the most inconsistent in speech rate. Ann and Beth spoke the fastest at 20 words per 5 seconds. They have a more consistent rate of talking throughout their whole stories.

There are also parts toward the end of the story where Chris got excited and enthusiastic. He spoke quickly. Here, he did not use a lot of fillers. All these conclusions are evident in his emotion, tone, mood, pace and rhythm. Words like um, uh and like occurred the most in Chris’s speech – 5 per every 100 words. Interestingly, Beth used only 1 filter word in every 100 words and Ann used less than 1 per every 300 words. However, Ann said like 8 times per every 100 words.

When comparing these three people, Chris’s amount of filler words seems extremely high. In reality, the American adult English speaker uses 7 fillers per every 100 words [1], although, we hardly notice them. It is obvious that fillers affect the speed of speaking, but they are not necessarily influenced by emotions. This project is not intended to catch all the errors in spoken language. Rather, it is attempting to reveal how we communicate verbally. I created an objective and subjective dissection of spoken language. Even though I subjected three narratives to this analysis, the system can apply to any kind of storytelling, speech or interview. It, metaphorically speaking, creates a blueprint that to parallel the story. It’s a system that is based a particular story, but it alternatively allows the reader to interpret the story on their own. Instead of seeing a story from one perspective in linear form, it gives readers another perspective on the stories. By looking at the system, you can better understand the tone, emotion and rhythms of the stories as well as observe the way people tell them.

Above all, I am interested in seeing the visual patterns created when language is filtered through a typographic system. There’s an element of beauty and surprise that comes from the organic texture of speech and language meeting an abstract system.

[1] Um. Slips, stumbles, and verbal blunders, and what they mean by Micheal Erard, Pantheon 2007