We present a method for improving the perceived naturalness of corpus-based speech synthesizers. It consists in removing pronounced pitch peaks in the original recordings, which typically lead to noticeable discontinuities in the synthesized speech. We perceptually evaluated this method using two concatenative and two HMM-based synthesis systems, and found that using it on the source recordings...

Modern theories of human evolution converge on the belief that our brain has been designed to cope with problems that occurred intermittently in our evolutionary past. Evidence suggests that, during over 99 percent of the evolutionary process leading to the emergence of our species, our ancestors communicated in a synchronous and colocated manner, and employing facial expressions, body language...

We present a hybrid approach to improving naturalness of electrolaryngeal (EL) speech while minimizing degradation in intelligibility. An electrolarynx is a device that artificially generates excitation sounds to enable laryngectomees to produce EL speech. Although proficient laryngectomees can produce quite intelligible EL speech, it sounds very unnatural due to the mechanical excitation produ...

For language learners it can be very difficult to speak without a non-native accent. This is due to the phenomenon called language transfer. At times, this can decrease the intelligibility of the speech and make it difficult to convey the correct message to the listener. Thus, many learners have a desire to speak more naturally in order not to be misjudged due to their pronunciation. For this, ...

This presentation reports work being done at SICS on novel interface techniques to large projection environments intended for multi-user applications. We have been exploring the notion of “naturalness” as it relates to interfaces. In this presentation we attempt to define this notion of naturalness and provide some reasons why the interaction media we present here meets that definition. We are ...

This paper proposes an architecture for an open-domain conversational system and evaluates an implemented system. The proposed architecture is fully composed of modules based on natural language processing techniques. Experimental results using human subjects show that our architecture achieves significantly better naturalness than a retrieval-based baseline and that its naturalness is close to...

in this research, different english translations of sa‘di’s bustan were studied. an anecdote was selected randomly with its three english translations to identify whether or not the translators have managed to convey the messages of the original poem. the three selected translations were examined according to two of the criteria that larson (1984) has proposed (accuracy and naturalness) for tes...

The paper elaborates on understanding, measuring and optimizing perceived color quality of natural images. We introduce a model for optimal color reproduction of natural scenes which is based on the assumption that color quality of natural images is constrained by perceived naturalness and colorfulness of these images. To verify the model, a few experiments were carried out where subjects estim...

A large number of extensions of Feature Oriented Domain Analysis (FODA) Feature Diagrams were introduced to compensate for a purported ambiguity and lack of precision and expressiveness of the original FODA feature diagrams (OFD). However, they never received a formal semantics, which is the hallmark of precision and unambiguity. We propose here a formal semantics for all these diagrams, thanks...

The relative contributions of segmental versus prosodic factors to the perceived naturalness of synthetic speech was measured by transplanting prosody between natural speech and the output of a diphone synthesizer. A small corpus was created containing matched sentence pairs wherein one member of the pair was a natural utterance and the other was a synthetic utterance generated with diphone dat...