Speech-based interfaces have great potential but are hampered by problems related to spoken language such as variability, noise and ambiguity. Speech Graffiti was designed to address these issues via a structured, universal interface protocol for interacting with simple machines. Since Speech Graffiti requires that users speak to the system in a certain way, we were interested in how users migh...

Speech is a natural mode to interact with others. With speech, we can express our words to others. Speech recognition is a way or technology where the statements or commands of human speech to understand and react accordingly. Speech recognition allows machining system to turn the incoming speech signals into commands through the process of identifying and understanding. It also creates the nat...

Speech synthesizers are optimized for fluent natural text. However, in a speech to speech translation system, they have to process machine translation output, which is often not fluent. Rendering machine translations as speech makes them even harder to understand than the synthesis of natural text. A speech synthesizer must deal with the disfluencies in translations in order to be comprehensibl...

In recent years, deep learning has achieved great success in speech enhancement. However, there are two major limitations regarding existing works. First, the Bayesian framework is not adopted in many such deep-learning-based algorithms. In particular, the prior distribution for speech in the Bayesian framework has been shown useful by regularizing the output to be in the speech space, and thus...

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the co...

This paper presents a statistical natural language generation scheme for trainable speech-to-speech machine translation (MT) systems for limited domain applications using a cascaded approach. The natural language generation scheme in the translation systems is based on a maximum entropy (ME) statistical model fully trained from a corpus, allowing flexible translation outputs. In this paper, the...

Human speech universally exhibits a 3- to 8-Hz rhythm, corresponding to the rate of syllable production, which is reflected in both the sound envelope and the visual mouth movements. Artificial perturbation of the speech rhythm outside the natural range reduces speech intelligibility, demonstrating a perceptual tuning to this frequency band. One theory posits that the mouth movements at the cor...

There is a long debate on the relative importance of spectral and temporal cues in speech perception theories. On the one hand, the highly-intelligible sine-wave speech (SWS) has been viewed as a representation of the global spectral structure of the speech signal. On the other hand, there is accumulating evidence showing that the temporal aspects of speech without spectral details provide suff...

We describe our Finnish audio-visual speech synthesizer, its evaluation and discuss possible improvements. We have combined a three dimensional facial model with a commercial audio text-to-speech synthesizer. The visual speech is based on a letter-to-viseme mapping and the animation is created by linear interpolation between the visemes. An intelligibility test was run to quantify the benefit o...