Semantics-Aware Human Motion Generation from Audio Instructions
This addresses the problem of weak semantic alignment in audio-conditioned motion generation for interactive technologies, representing a novel task rather than an incremental improvement.
The paper tackles the problem of generating human motions that align with the semantics of audio instructions, rather than just rhythms, by proposing an end-to-end framework using a masked generative transformer with a memory-retrieval attention module. The result shows that audio instructions can convey semantics similar to text, enabling more practical and user-friendly interactions.
Recent advances in interactive technologies have highlighted the prominence of audio signals for semantic encoding. This paper explores a new task, where audio signals are used as conditioning inputs to generate motions that align with the semantics of the audio. Unlike text-based interactions, audio provides a more natural and intuitive communication method. However, existing methods typically focus on matching motions with music or speech rhythms, which often results in a weak connection between the semantics of the audio and generated motions. We propose an end-to-end framework using a masked generative transformer, enhanced by a memory-retrieval attention module to handle sparse and lengthy audio inputs. Additionally, we enrich existing datasets by converting descriptions into conversational style and generating corresponding audio with varied speaker identities. Experiments demonstrate the effectiveness and efficiency of the proposed framework, demonstrating that audio instructions can convey semantics similar to text while providing more practical and user-friendly interactions.