基于广电网络的人工智能语音能力系统建设  

Construction of Artificial Intelligence Speech Capability System Based on Radio and Television Network

作　　者：胡俊 占亿民 冒海波 王世祥 潘彩霞 Hu Jun;Zhan Yimin;Mao Haibo;Wang Shixiang;Pan Caixia(JSCN Institute of Technology Corporation Limited,Jiangsu 210001,China)

机构地区：[1]江苏有线技术研究院有限公司,江苏210001

出　　处：《广播与电视技术》2021年第1期75-79,共5页Radio & TV Broadcast Engineering

摘　　要：当前,语音识别、自然语言处理、图像识别、深度学习等为代表的人工智能技术飞速发展,推动着广电行业新技术、新产品、新业态的不断迭代。江苏有线通过融合人工智能及语音识别等技术,聚焦用户痛点,持续创新,建设了一套人工智能语音能力系统,利用智能语音能力为各类传统电视业务赋能。At present, rapid development of artificial intelligence technologies, such as speech recognition, natural language processing, image recognition and deep learning, promotes continuous iteration of new technologies, new products and new formats in radio and television industry. JSCN focuses on users’ pain points, makes continuous innovation, builds a set of artificial intelligence speech capability system by integrating artificial intelligence and speech recognition technologies, which enables all kinds of traditional TV services with intelligent speech capability.

关 键 词：有线网络 人工智能 语音识别 

分 类 号：TN941.3[电子电信—信号与信息处理]