Text-to-motion generation allows language-driven animation, yet current models struggle to deliver long-range coherence and fine-grained limb coordination. A competitive system must (i) preserve temporal consistency across hundreds of frames, (ii) synchronize limb motions, and (iii) align nuanced sentences with a spectrum of plausible trajectories. We introduce BiPO, the first part-based bidirectional autoregressive network trained with a lightweight Partial Occlusion regulariser. Each limb attends to both past and future frames for anticipatory coordination, while stochastic masking weakens spurious cross-part dependencies and encourages varied solutions. On HumanML3D and KIT-ML, BiPO lowers FID by 15–30% relative to MoMask and BAMM, secures the highest human-perceived realism scores, and sets new state-of-the-art results on motion-editing tasks requiring infill from partial sequences. These findings demonstrate that bidirectional reasoning coupled with Partial Occlusion yields a length-agnostic, high-fidelity framework for expressive, language-conditioned motion synthesis.

This paper proposes a novel generative model called PUGAN, which progressively synthesizes high-quality audio in a raw waveform. Progressive upsampling GAN (PUGAN) leverages the progressive generation of higher-resolution output by stacking multiple encoder-decoder architectures. Compared to an existing state-of-the-art model called WaveGAN, which uses a single decoder architecture, our model generates audio signals and converts them to a higher resolution in a progressive manner, while using a significantly smaller number of parameters, e.g., 3.17x smaller for 16 kHz output, than WaveGAN. Our experiments show that the audio signals can be generated in real time with a comparable quality to that of WaveGAN in terms of the inception scores and human perception.

Onomatopoeia refers to a word that phonetically imitates, resembles the sound, or depict an event at hand. In languages like Korean and Japanese, it is used in everyday conversation to emphasize certain situation and enrich the prose. In this poster, we explore if the use of onomatopoeia, visualized and added to the usual sound feedback, could be taken advantage to increase or alter the perceived realism of the sound feedback itself, and furthermore of the situation at hand in virtual reality. A pilot experiment was run to compare the user's subjective perceived realism and experience under four test conditions of presenting a simple physical interaction, accompanying it with: (1) just the "as-is" sound (baseline), (2) "as-is" sound and onomatopoeia, (3) a representative sound sample (e.g. one for all different collision conditions), and (4) a representative sound sample and onomatopoeia. Our pilot study has found that the use of onomatopoeia can alter and add on to the perceived realism/naturalness of the virtual situation such that the experiences of the single representative sound added with the onomatopoeia and "as-is" sound were deemed similar.