Kindly inform Stella to bring along the following items from the store: six spoons of fresh snow peas, five thick slabs of blue cheese, and perhaps a snack for her sibling Bob. In addition, we require a small plastic snake and a large toy frog for the children. Stella can pack these items into three red bags, and we will meet her at the train station on Wednesday.
The Online Speech Accent Archive, located at George Mason University in Fairfax, Virginia, contains over 1,300 unique recordings of individuals speaking the aforementioned passage in various accents. The purpose of the archive is to showcase a broad spectrum of speech accents and dialects from diverse linguistic backgrounds around the world. The collection comprises of both native and non-native English speakers who were selected as they correctly articulate most of the consonants, vowels, and clusters of Standard American English.
Dr. Steven H Weinberger, a linguistics professor and the archive director, arranges for the collection of samples from individuals worldwide. Though some recordings are of low quality and are discarded, the archive accepts a vast majority that meets its quality standard. Contributors also provide information about their demographics and linguistic background as it helps users identify key variables of each accent.
The archive allows for language and geographical sorting, and a phonetic transcription accompanies each recording. These features permit users to compare accents of speakers from different regions and backgrounds. Listeners can appreciate the differences and similarities among the various accents.
Dr. Weinberger asserts that accents are not a result of faulty learning but are systematic to an individual’s native language. As children, we typically develop an automatic ability to differentiate between a speaker from our community and a foreign one. The speakers of a particular foreign language share a common sound system, such that the addition, deletion, substitution, or alteration of certain sounds by such speakers is predictable and systematic to their native language. French speakers, for instance, have an accent in English because they apply their French sound system when communicating in English.
Despite biases against certain accents, the archive aims to increase appreciation of different speech sounds by society. It is critical to bear in mind that accents should not be dismissed as erroneous or incorrect pronunciation. Nevertheless, the study of accents reflects upon the broader aspects of speech, which are rich and complex.
The archives provide further evidence supporting the idea of a critical period. Only young individuals are capable of speaking English natively. One may take a Korean learner as an example. Suppose they initiate learning English at 11 and has spent 20 years of their life in the USA while speaking English, they will still exhibit the Korean accent. However, an individual who commences learning English at four and shifts to the USA for five years will not hold a Korean accent. Therefore, it is the age of starting to learn, not the exposure duration, that plays a vital role.
The archive serves both educational and research purposes. Numerous groups utilize it, including linguists and phoneticians, English teachers to non-native speakers, speech pathologists, and engineers who train speech recognition mechanisms. Although Weinberger might become tired of hearing endlessly about fresh snow peas and blue cheese, plenty of individuals appreciate it. A quick Google search reveals that people leverage it for many projects, from ringtones to art projects. YouTube is full of recordings, and Irish composer Cathal Roche has created saxophone pieces based on the archive.
Although there have been academic papers and master’s theses based on the archive, Weinberger thinks that most people take pleasure in listening to accented speech sounds. In his words, "I’m sure there are plenty of drinking games based on the archive!"
Actors who struggle to get a given accent right might have been helped for those who had the archive at that time. Examples include Van Dyke and Sean Connery, who topped Empire magazine’s poll for the worst accents in cinema history for their work in Mary Poppins (1964) and The Untouchables (1987), respectively. Actors frequently contact Weinberger to express their gratitude for helping them with obscure speech accents for scripts.
Of what’s next, Weinberger says they are preparing for a significant overhaul, which will bring better maps via Google, more searchable sounds, and more phonetic inventories from the world’s languages. They are compiling a database on the syllable structures available in the world’s languages. Additionally, they have created a computational device that can automatically compare two accents and show the specific phonological speech patterns that distinguish one accent from another. The best part? It’s free to anyone.
Weinberger always carries a voice recorder with him to capture new accents whenever he encounters one. Although he doesn’t know how long the archive will run, he explains that they only represent 300 native languages while over 6,000 languages exist worldwide. They still have a long way to go.
