Flexible Speech Synthesis Using Weighted Finite State Transducers

Two-way finite state transducers are considered with a fixed number of pebbles, of which the life times are nested. In the deterministic case, the transductions computed by such pebble transducers are closed under composition, and they can be realized by the composition of one-pebble transducers. The ranges of the k-pebble transducers form a hierarchy with respect to k, their finiteness problem...

− At PTB, one designed and investigated measurement setups to determine the deformation of force transducers while applying a static or dynamic force. To measure the deformation, a differential vibrometer [1] is used which is integrated in the measurement setups. Using the deformation results one is able to calculate the transducers stiffness. An evaluation of the measurement uncertainty is pre...

Software which corrects the dynamic error of force transducers in impact force measurements using their own output signal has been developed. The software corrects the output waveform of the transducers using the output waveform itself, estimates its uncertainty and displays the results. In the experiment, the dynamic error of three transducers of the same model are evaluated using the Levitati...

Introduced by Dr. Charles Kelman in 1962, phacoemulsification machines have undergone constant improvement, ever increasing both their complexity and safety. There is one principle, however, that remains unchanged. All phaco machines consist of a computer to generate electrical signals and a transducer to turn these electronic signals into mechanical energy. The energy thus produced is passed t...

This paper describes the construction of a system for tactile sound using tactile transducers and explores the compositional potential for a system of these characteristics. It provides a short overview of the technologies involved and the composition methods and studio techniques developed to implement this project. Sense for fourteen discrete channels includes a surround 5.1-sound composition...

We present two simple finite-state transducer based strategies for tweet normalization. One relies on hand-written correction rules designed to capture commonly occurring misspellings and abbreviations, while the other tries to automatically induce an error model from a gold standard corpus of normalized tweets.