In speaker recognition applications, the changes of emotional states are main causes of errors. The ongoing work described in this contribution attempts to enhance the performance of automatic speaker recognition (ASR) systems on emotional speech. Two procedures that only need a small quantity of affective training data are applied to ASR task, which is very practical in real-world situations. ...

We present a framework for handling emotional variations in a speech-based natural language system for use in the MRE virtual training environment. The system is a first step toward addressing issues in emotion-based modeling of verbal communicative behavior. We cast the problem of emotional generation as a distance minimization task, in which the system chooses between multiple valid realizati...

An emotional speech corpus of Finnish was collected that includes utterances of four emotional states of speakers. More than 40 prosodic features were derived and automatically computed for the speech samples. Statistical classification experiments with kNN classifier and human listening tests indicate that emotion recognition performance comparable to human listeners can be achieved.

An emotional speech corpus of Finnish was collected that includes utterances of four emotional states of speakers. More than 40 prosodic features were derived and automatically computed for the speech samples. Statistical classification experiments with kNN classifier and human listening tests indicate that emotion recognition performance comparable to human listeners can be achieved.

In the present study, the vocal expressions of the emotions anger, happiness, fear, boredom and sadness are acoustically analyzed in relation to neutral speech. The emotional speech material produced by actors is investigated especially with regard to spectral and segmental changes which are caused by different articulatory behavior accompanying emotional arousal. The findings are interpreted i...

In this paper we propose a new approach for analysis of emotional speech prosody features. The aim of the analysis is definition of emotional features that characterise emotions. Analysis was performed on emotional speech databases that were recorded in the framework of the project "Multimodal Analysis/Synthesis System for Human Interaction to Virtual and Augmented Environments" (Interface). Th...

The goal of this paper is to assess the effect of emotional state of a speaker when text-independent speaker identification is performed. Mel-frequency cepstral coefficients are the features of the speech signal used for speaker recognition. For training the speaker models and testing the system, Support Vector Machines are employed. Berlin emotional speech database, which contains 10 different...

An emotional speech corpus of Finnish was collected that includes utterances of four emotional states of speakers. More than 40 prosodic features were computed for speech samples. Statistical classification experiments indicate that emotion recognition performance comparable to human listeners can be achieved. However, due to the small amount of data, the results must be interpreted cautiously.

In our fully automatic corpus-based method of generating fundamental frequency (F0) contours for emotional speech synthesis, an improvement was realized related to the process of corpus preparation. The method assumes the generation process model and predicts its command parameters using binary regression trees with inputs of linguistic information of the sentence to be synthesized. Because of ...

BACKGROUND The study of emotional speech perception and emotional prosody necessitates stimuli with reliable affective norms. However, ratings may be affected by the participants' current emotional state as increased anxiety and depression have been shown to yield altered neural responding to emotional stimuli. Therefore, the present study had two aims, first to provide a database of emotional ...