Attempts to add emotion effects to synthesised speech have existed for more than a decade now. Several prototypes and fully operational systems have been built based on different synthesis techniques, and quite a number of smaller studies have been conducted. This paper aims to give an overview of what has been done in this field, pointing out the inherent properties of the various synthesis techniques used, summarising the prosody rules employed, and taking a look at the evaluation paradigms. Finally, an attempt is made to discuss interesting directions for future development.

The article describes a database of emotional speech. Ten actors (5 female and 5 male) simulated the emotions, producing 10 German utterances (5 short and 5 longer sentences) which could be used in everyday communication and are interpretable in all applied emotions. The recordings were taken in an anechoic chamber with high-quality recording equipment. In addition to the sound electro-glottograms were recorded. The speech material comprises about 800 sentences (seven emotions * ten actors * ten sentences + some second versions). The complete database was evaluated in a perception test regarding the recognisability of emotions and their naturalness. Utterances recognised better than 80 % and judged as natural by more than 60 % of the listeners were phonetically labelled in a narrow transcription with special markers for voice-quality, phonatory and articulatory settings and articulatory features. The database can be accessed by the public via the internet

We present a data-mining experiment on feature selection for automatic emotion recognition. Starting from more than 1000 features derived from pitch, energy and MFCC time series, the most relevant features in respect to the data are selected from this set by removing correlated features. The features selected for acted and realistic emotions are analysed and show significant differences. All features are computed automatically and we also contrast automatically with manually units of analysis. A higher degree of automation did not prove to be a disadvantage in terms of recognition accuracy. 1.

Proceedings

Communication is an essential part of our life. Though, not only communication is the key – it is all about emotional (prosodic) communication. Due to empirical research, people, who are augmentative communicators and speak with a voice output communication aid, want to express their emotions in the same way as everybody else – it is one of their deepest interests (Portnuff, 2006; Hoffmann and Wülfing, 2010). So far, current devices lack the opportunity of emotional utterances. This circumstance leads not only to a huge usability deficit, but furthermore, it is an obstacle to develop emotional competence and to behave as well as regulate one´s emotion adequately (Blackstone and Wilkins, 2009). This article aims to increase the sensitivity for the importance of emotional communication. Furthermore, it tries to give first suggestions for implementing an usable device that supports users with a voice output communication aid to express emotional utterances. This could be done by using phrase-generation, as mentiond by Vanderheyden and Pennigton (1998). 1

The goal of this MSc project was to build unit selection voice that could portray emo-tions in various intensities. A suitable definition of emotion was developed along with a descriptive framework that supported the work carried out. Two speakers were recorded portraying happy and angry speaking styles, additionally a neutral database was also recorded. One voice was built for each speaker that included all the speech from that speaker. A target cost function was implemented that choses units from the database according to emotion mark-up in the database. The Dictionary of Affect [30] supported the emotional target cost function by providing an emotion rating for words in the target utterance. If a word was particularly emotional, units from that emotion were favoured. In addition intensity could be varied which resulted in a bias to select more emotional units. A perceptual evaluation was carried out and subjects were able to recognise reliably, emotions with varying amounts of emotional units present in the target utterance.

Voice quality, defined by John Laver as the characteristic auditory colouring of a speaker’s voice, is a significant feature of speech, and it is used to signal various properties such as emotions, intentions, and mood of the speaker. While voice quality measurement techniques and algorithms have been developed, much work is needed to obtain a comprehensive view of the function and analysis of human voice in the production of different voice qualities. Two major research questions are presented in this thesis: First, how can the most important laryngeal voice quality features be analyzed, and second, how do the voice quality features affect different facets of vocal expression? To answer these questions, five separate studies of the analysis methodology and two studies regarding the voice quality behaviour were