摘 要
从键盘到多点触摸再到语音交互,语音技术已经广泛应用于市场上的APP产品。而Siri的出现更是让人们见到了语音助手的优点和便捷之处,打电话、发信息、找应用等等操作,只要一个语音口令就可以。
讯飞开放平台是智能交互技术的服务平台,为开发者提供各种服务,包括语音基础能力类的合成、识别,语音定制服务类的语音唤醒、开放语义和语音云,模式识别类的人脸识别,开放统计类的移动应用分析等等。作为全球首个智能语音交互平台,让广大开发者在开发过程中获益匪浅。
本应用就是基于科大讯飞的开放平台开发的,使用它的MSC(Mobile Speech Client,移动语音终端)Android版SDK,应用开放语义、语音合成服务的接口,将用户的语音信息传到云端,分析语音的意图,给出对应的回答。并将回答以JSON形式返回到Android手机,在手机端经过解析JSON数据后,将语义解析结果利用科大讯飞的SDK中的在线语音合成功能将结果说出。从而实现语义识别、智能问答、垂直搜索,达到释放双手、人机智能交互的目的。
关键词:科大讯飞;开放语义;人机交互
From the keyboard to the multi touch and voice interaction, speech technology has been widely used in APP products on the market. The emergence of Siri let people to see the advantages of voice assistant and convenience, phone calls, send a message, find applications and so on, just a voice command .
IFLYTEK intelligent open platform is interactive technology services platform, providing various kinds of service for developers, including voice synthesis, recognition of basic ability, speech customization service of semantic and phonological wake up, open cloud, face recognition pattern recognition, statistical analysis of the open cloud, open cloud, face recognition pattern recognition, statistical analysis of the open mobile application and so on. As the world's first intelligent interactive platform, so that the majority of developers in the development process of benefit.
This application is based on iFLYTEK open platform, using the MSC ( Mobile Speech Client,the mobile terminal ) SDK for Android, use the open service interfacesemantics, speech synthesis, speech user information to the cloud, analysis of speechintention, give out the corresponding answer. And the answer in the form of JSON to return to the Android mobile phone, mobile phone in the end through the analysis of JSON data, the semantic parsing results using online speech synthesis function iFLYTEK will result in SDK say. In order to achieve the semantic recognition, intelligent question answering, vertical search, to release the hands of human-computer interaction, the purpose of.
Keywords: iFLYTEK ; open semantic; human-computer interaction