sonic_signatures/

Summer 2017 research into visual text analysis of sonic signatures/

Sonic signatures is an exploration of the relationship between the sound of words and the people who speak them. It is a collection of data visualization and analysis techniques to explore the kinds of vowels an consonants that show up in a text and in what concentrations. Sonic signatures has its origins as a technique to explore whether there are similarities in the way that certain Shakespeare characters sound - do villains use lots of phonemes that cause one to bear one’s teeth when spoken? Do rich old men use a higher concentration of “dark” sounding vowels? This technique will hopefully be extendable to other bodies of text as the project progresses.

Getting Started

The foundational process of our software is using nltk to convert texts from American English to phonemes, which are then categorized and analyzed by frequency.

With Shakespeare, the texts went from res to dest using phonetic_transcriber.py. The phonetic_transcript method(self, file_name) will do the same for another file.

Prerequisites

NLTK - Can be installed using instructions at https://www.nltk.org/install.html

bs4 - https://www.crummy.com/software/BeautifulSoup/bs4/doc/#quick-start

Built With

• Flask - The web framework used

Authors

• Eric Alexander - Supervision - EAlexander

• Estelle Bayer - EstelleEvelyn

• Liz Nichols - nicholsl

See also the list of contributors who participated in this project.

License

This project is licensed under

Acknowledgments

• Inspired by University of Utah's Poemage