System

A system comprising:

• circuitry configured to:
• receive, via a communication interface operating over a packet-switched network, audio information, and preprocess the audio information by performing noise reduction processing and amplitude normalization;
• extract frame-based feature vectors from the audio information and input the feature vectors to an acoustic model comprising convolutional layers and self-attention layers to generate phonetic representations, and combine the phonetic representations with a language model using beam search decoding to generate character information;
• extract prosodic feature quantities comprising fundamental frequency, intensity, speech rate, and pause duration from the audio information, compute statistical descriptors for each prosodic feature quantity over analysis windows, and classify the prosodic feature quantities using an emotion classifier comprising at least one of a long short-term memory network and a transformer-based classifier to specify emotion information;
• generate a sign language intermediate representation comprising a sequence of sign units, grammatical annotations, non-manual markers, and a target sign language type on a basis of the character information and the emotion information;
• construct a prompt sentence including the sign language intermediate representation, the emotion information, and the target sign language type, supply the prompt sentence to a generative AI model, and obtain structured sign language expression data comprising hand shape, position, movement, duration, and facial expression tags;
• convert the structured sign language expression data into an animation script comprising skeletal information, joint angle information, and facial expression information, and apply motion smoothing processing to the joint angle information;
• and
• transmit the animation script to a user terminal via the packet-switched network for three-dimensional rendering of a virtual character with user-configurable appearance and motion characteristics.

A method performed by circuitry of a system, the method comprising:

• receiving, via a communication interface coupled to a packet-switched network, audio information;
• analyzing the audio information to generate character information;
• extracting prosodic feature quantities from the audio information and classifying the prosodic feature quantities to specify emotion information;
• generating a visual expression representation on a basis of the character information and the emotion information;
• and
• converting the visual expression representation into animation data and transmitting the animation data to a user terminal via the packet-switched network.