Greg Kochanski

Learn More
We explored a database covering seven dialects of British and Irish English and three different styles of speech to find acoustic correlates of prominence. We built classifiers, trained the classifiers on human prominence/nonprominence judgments, and then evaluated how well they behaved. The classifiers operate on 452 ms windows centered on syllables, using(More)
This paper describes a novel prosody generation model. We intend it to broadly support many linguistic theories and multiple languages, for the model imposes no restriction on accent categories and shapes. This capability is crucial to the next generation of text-to-speech systems that will need to synthesize intonation variations for different speech acts,(More)
We describe models of Mandarin prosody that allow us to make quantitative measurements of prosodic strengths. These models use Stem-ML, which is a phenomenological model of the muscle dynamics and planning process that controls the tension of the vocal folds, and therefore the pitch of speech. Because Stem-ML describes the interactions between nearby tones,(More)
We model the differences between declarative and interrogative intonation in Chinese with Stem-ML, an intonation description language combined with an algorithm for translating tags into quantitative prosody. Our study shows that the diverse surface patterns can be accounted for by two consistent gestures: 1. Interrogative intonation has a higher phrase(More)
The mathematical models of intonation used in speech technology are often inaccessible to linguists. By the same token, phonological descriptions of intonation are rarely used by speech technologists, as they cannot be implemented directly in applications. Consequently, these research communities do not benefit much from each other's insights. In this(More)
Although the pitch of the human voice is continuously variable, some linguists contend that intonation in speech is restricted to a small, limited set of patterns. This claim is tested by asking subjects to mimic a block of 100 randomly generated intonation contours and then to imitate themselves in several successive sessions. The produced f0 contours(More)
Signal detection theory (SDT) makes the frequently challenged assumption that decision criteria have no variance. An extended model, the Law of Categorical Judgment, relaxes this assumption. The long accepted equation for the law, however, is flawed: It can generate negative probabilities. The correct equation, the Law of Categorical Judgment (Corrected),(More)
“Hackers” have written malicious programs to exploit online services intended for human users. As a result, service providers need a method to tell whether a web site is being accessed by a human or a machine. We expect a parallel scenario as spoken language interfaces become common. In this paper, we describe a Reverse Turing Test (i.e., an algorithm that(More)