Funding, ideology and the aesthetics of the development film in postcolonial Zimbabwe

This article considers the relationships between aesthetics and ideology in donor-funded ‘development’ film-making from Zimbabwe, examining in particular how the films’ producers have attempted to popularize a genre of film-making that has its roots in colonial cinema. Making close reference to two productions from the Harare-based Media for Development Trust (MFD) – Neria (Godwin Mawaru, 1992), and Everyone’s Child (Tsitsi Dangarembga, 1996) (both of which may be regarded as archetypal examples of their genre) – the article demonstrates how the films deploy a range of aesthetic strategies to imbue a set of narratives drawn from colonial development films with greater impact and cultural resonance for contemporary local audiences. The article also suggests that close analysis of these strategies may provide insights into the relationships between the films’ aesthetic dimensions and wider ideological issues in the region.

On-line Emotion Recognition in a 3-D Activation-Valence-Time Continuum using Acoustic and Linguistic Cues

For many applications of emotion recognition, such as virtual agents, the system must select responses while the user is speaking. This requires reliable on-line recognition of the user’s affect. However most emotion recognition systems are based on turnwise processing. We present a novel approach to on-line emotion recognition from speech using Long Short-Term Memory Recurrent Neural Networks. Emotion is recognised frame-wise in a two-dimensional valence-activation continuum. In contrast to current state-of-the-art approaches, recognition is performed on low-level signal frames, similar to those used for speech recognition. No statistical functionals are applied to low-level feature contours. Framing at a higher level is therefore unnecessary and regression outputs can be produced in real-time for every low-level input frame. We also investigate the benefits of including linguistic features on the signal frame level obtained by a keyword spotter.