In recent years, the performance of speech synthesis systems has been improved thanks to deep learning-based models, but generating expressive audiovisual speech is still an open issue. The variational auto-encoders (VAE)s are recently proposed to learn latent representations of data. In this paper, we present a system for expressive text-to-audiovisual speech synthesis that learns a latent embedding space of emotions using a conditional generative model based on the variational auto-encoder framework. When conditioned on textual input, the VAE is able to learn an embedded representation that captures emotion characteristics from the signal, while being invariant to the phonetic content of the utterances. We applied this method in an unsupervised manner to generate duration, acoustic and visual features of speech. This conditional variational auto-encoder (CVAE) has been used to blend emotions together. This model was able to generate nuances of a given emotion or to generate new emotions that do not exist in our database. We conducted three perceptive experiments to evaluate our findings.

In this study, the precision of markerless acquisition techniques have been assessed when used to acquire articulatory data for speech production studies. Two different markerless systems have been evaluated and compared to a marker-based one. The main finding is that both markerless systems provide a reasonable result during normal speech and the quality is uneven during fast articulated speech. The quality of the data is dependent on the temporal resolution of the markerless system.

HAL is a multi-disciplinary open access archive for the deposit and dissemination of scientific research documents, whether they are published or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers. L'archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d'enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.

In the context of developing an expressive audiovisual speech synthesis system, the quality of the audiovisual corpus from which the 3D visual data will be extracted is important. In this paper, we present a perceptive case study on the quality of the expressiveness of a set of emotions acted by a semi-professional actor. We have analyzed the production of this actor pronouncing a set of sentences with acted emotions, during a human emotion-recognition task. We have observed different modalities: audio, real video, 3D-extracted data, as unimodal presentations and bimodal presentations (with audio). The results of this study show the necessity of such perceptive evaluation prior to further exploitation of the data for the synthesis system. The comparison of the modalities shows clearly what the emotions are, that need to be improved during production and how audio and visual components have a strong mutual influence on emotional perception.

Great improvement has been made in the field of expressive audiovisual Textto-Speech synthesis (EAVTTS) thanks to deep learning techniques. However, generating realistic speech is still an open issue and researchers in this area have been focusing lately on controlling the speech variability. In this paper, we use different neural architectures to synthesize emotional speech. We study the application of unsupervised learning techniques for emotional speech modeling as well as methods for restructuring emotions representation to make it continuous and more flexible. This manipulation of the emotional representation should allow us to generate new styles of speech by mixing emotions. We first present our expressive audiovisual corpus. We validate the emotional content of this corpus with three perceptual experiments using acoustic only, visual only and audiovisual stimuli. After that, we analyze the performance of a fully connected neural network in learning characteristics specific to different emotions for the phone duration aspect and the acoustic and visual modalities. We also study the contribution of a joint and separate training of the acoustic and visual modalities in the quality of the generated synthetic speech. In the second part of this paper, we use a conditional variational auto-encoder (CVAE) architecture to learn a latent representation of emotions. We applied this method in an unsupervised manner to generate features of expressive speech. We used a