Chinese sign language synthesis system driven by markup language

Abstract

The invention discloses a Chinese sign language synthesis system driven by a markup language. The system comprises two modules, namely a sign language rhythm controller and a sign language behavior synthesizer, and a Chinese sign language markup language serves as a system input data interface and a data interface between the modules. By adoption of the system, standardized and visual Chinese sign language description can be provided. The system accords with a Chinese sign language grammar specification and is high in extensibility.

Description

technical field [0001] The invention relates to the technical field of digital home, in particular to a mark language-driven Chinese sign language synthesis system. Background technique [0002] With the continuous development of human-computer interaction technology, highly realistic virtual images (Embodied Agent) are widely used in various interactive fields due to their lower manpower consumption, more flexible production and editing methods, and promote the development of human-computer interaction interfaces. Towards a more friendly and natural way, sign language synthesis is one of the important applications. Because deaf people have reading barriers to written texts created by spoken language grammar, their reception efficiency of information is much lower than that of hearing people. Therefore, the current research on sign language synthesis mainly focuses on the real-time translation and synthesis technology from natural texts to sign language animations. The func...

AI Technical Summary

Problems solved by technology

[0008] (1) Lack of Chinese sign language features to support the existing sign language description language does not have the ability to universally apply to various languages, and is not applicable to some special phenomena and composition structures in Chinese sign language

[0009] (2) Lack of prosodic marks The existing sign language description language is based on the translation of literal content and lacks descriptions of semantic and prosodic features

[0010] (3) Lack of high-level marker definition The existing sign language description language mainly works on the data interface of the speech layer, although the direct definition of the behavior composition at the speech layer can provide an effective detailed description of the expression, such as the phoneme composition of gestures and the movement of other parts of the body, However, it will limit the diversity and naturalness of synthetic effects to a certain extent, resulting in rigid and mechanized virtual human movements.

In addition, this design structure cannot effectively support the generation mechanism of the prosodic model, and requires a higher level of knowledge for manual editing

Images