Building Autonomous Sensitive Artificial Listeners

This paper describes a substantial effort to build a real-time interactive multimodal dialogue system with a focus on emotional and non-verbal interaction capabilities. The work is motivated by the aim to provide technology with competences in perceiving and producing the emotional and non-verbal behaviours required to sustain a conversational dialogue. We present the Sensitive Artificial Listener (SAL) scenario as a setting which seems particularly suited for the study of emotional and non- verbal behaviour, since it requires only very limited verbal understanding on the part of the machine. This scenario allows us to concentrate on non-verbal capabilities without having to address at the same time the challenges of spoken language understanding, task modeling etc. We first report on three prototype versions of the SAL scenario, in which the behaviour of the Sensitive Artificial Listener characters was determined by a human operator. These prototypes served the purpose of verifying the effectiveness of the SAL scenario and allowed us to collect data required for building system components for analysing and synthesising the respective behaviours. We then describe the fully autonomous integrated real-time system we created, which combines incremental analysis of user behaviour, dialogue management, and synthesis of speaker and listener behaviour of a SAL character displayed as a virtual agent. We discuss principles that should underlie the evaluation of SAL-type systems. Since the system is designed for modularity and reuse, and since it is publicly available, the SAL system has potential as a joint research tool in the affective computing research community.

Artificial Impedance Surfaces for Reduced Dispersion in Antenna Feeding Systems

An impedance surface is presented that reduces the dispersion experienced upon propagation of broadband pulses within rectangular waveguides. The surface impedance is selected so that, within a frequency range, the transverse resonance condition is satisfied for longitudinal wavenumber that varies linearly with frequency. A synthesis procedure for practical surface topologies consisting of periodic dipole arrays is described. An example involving a finite structure is employed to illustrate the reduced dispersion. Numerical simulation results obtained from in-house mode-matching method as well as HFSS are presented. A prototype is fabricated and tested experimentally validating the theoretical predictions.

Kasstasin: A novel potent vasoconstrictor peptide from the skin secretion of the African red-legged running frog, Kassina maculata

Amphibian skin secretions are established sources of bioactive peptides. Here we describe the isolation, structural and pharmacological characterisation of a novel vasoconstrictor peptide from the skin secretion of the African hyperoliid frog, Kassina maculata, which exhibits no structural similarity to any known class of amphibian skin peptide. The peptide consists of 21 amino acid residues, FIKELLPHLSGIIDSVANAIK, and is C-terminally amidated. The provisional structure was obtained by MS/MS fragmentation using an Orbitrap mass spectrometer and L/I ambiguities were resolved following molecular cloning of biosynthetic precursor-encoding cDNA. A synthetic replicate of the peptide was found to possess weak antimicrobial and haemolytic activities but was exceptionally effective in constricting the smooth muscle of rat tail artery (EC50 of 25pM). In reflection of its exceptional potency in constricting rat arterial smooth muscle, the peptide was named kasstasin, a derivation of Kassina and “stasis” (stoppage of flow). These data illustrate the continuing potential of amphibian skin secretions to provide novel natural peptide templates for biological evaluation.