library_name: transformers
license: mit
pipeline_tag: text-classification
Model Description
We introduce ModCon-Task-Identifier, a fine-tuned BERT model that is capable of identifying the modality conversion task type based on the user’s prompt. For instance, if the user’s prompt is ‘read this text’, the model will output ‘TTS’ (Text-to-Speech). The model was developed as a part of the Multi-Agent MATE project, the goal of which is to develop a universal multi-agent modality conversion framework. Based on the user’s query, the system will convert the input file to the desired format by changing the modality (for instance, a text can be converted to an image, or a video can be converted to an audio)
The official project repository and the full project code are available at https://github.com/AlgazinovAleksandr/Multi-Agent-MATE
This model is based on the paper MATE: LLM-Powered Multi-Agent Translation Environment for Accessibility Applications