Real-time communication system for use by individuals with speech and hearing impairments, has automatic speech recognition module for processing spoken language and translating spoken language into text
	MISHRA A; SHARMA K; UPADHYAY A; JAIN N
	2025-03-28
专利权人	MISHRA A (MISH-Individual) ; SHARMA K (SHAR-Individual) ; UPADHYAY A (UPAD-Individual) ; JAIN N (JAIN-Individual)
申请日期	2025-03-28
专利号	IN202511029886-A
成果简介	NOVELTY - The system has a natural language processing (NLP) module for incorporating a Large Language Model (LLM) to convert recognized gestures into structured text or synthesized speech output using a text-to-speech (TTS) engine. An automatic speech recognition (ASR) module processes spoken language and translates the spoken language into text. A sign language generation (SLG) module converts text-based responses into animated sign language representations or visual cues for an impaired individual. An adaptive learning module utilizes reinforcement learning and user feedback to refine recognition accuracy based on personal signing styles, environmental variations and regional sign language dialects. A user interface is deployed on mobile devices, wearable smart devices or embedded assistive hardware to facilitate seamless bidirectional communication between sign language users and non-sign language users. USE - Real-time communication system for use by individuals with speech and hearing impairments. ADVANTAGE - The system enhances recognition accuracy by integrating computer vision and natural language processing adapting to various sign language dialects, and ensures fluid communication without requiring specialized training for users. The system can adapt to individual user styles, recognize contextual variations and provide an intuitive interface for seamless communication, and significantly improves accessibility, independence and social integration for individuals with communication impairments by bridging the gap between sign language users and non-users, and ensures natural and context-aware communication by integrating a large language model (LLM) to enhance translation accuracy.
IPC 分类号	G06N-003/045 ; G06N-003/08 ; G06V-040/20 ; G09B-021/00 ; G10L-013/00
国家	印度
专业领域	信息技术
语种	英语
成果类型	专利
文献类型	科技成果
条目标识符	http://119.78.100.226:8889/handle/3KE4DYBR/13421
专题	中国科学院新疆生态与地理研究所
作者单位	1.MISHRA A (MISH-Individual) 2.SHARMA K (SHAR-Individual) 3.UPADHYAY A (UPAD-Individual) 4.JAIN N (JAIN-Individual)
推荐引用方式 GB/T 7714	MISHRA A,SHARMA K,UPADHYAY A,et al. Real-time communication system for use by individuals with speech and hearing impairments, has automatic speech recognition module for processing spoken language and translating spoken language into text. IN202511029886-A[P]. 2025.

条目包含的文件
条目无相关文件。

除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。