商务传媒学院知识库

  随着媒介技术的不断更迭，“声音”作为辅助性媒介的地位正在发生改变，声音与音频的价值逐渐被挖掘出来。在移动信息传播的背景下，网络音频开始进入大众视野，各大音频平台的勃兴为音频行业带来了新的发展机遇。近年来，网络音频头部平台纷纷加入使用AI语音合成技术的行列，通过智能化、人性化、专业化服务，为听众带来不一样的听觉体验。而在听觉文化转向背景下，AI语音合成技术形塑了全新的人工声音景观，所以探究网络音频内容中AI声音要素的呈现很有必要。

  本研究首先对AI语音合成技术与网络音频平台进行溯源，发现深度学习方法是当前语音合成的主流方法，而AI语音合成技术在网络音频平台中的应用已进入蓬勃发展期，其实AI作品在视频平台早已有之，如今借助网络音频发展的东风，拥有了不可小觑的发展潜力。本文针对目前市场上具有代表性的网络音频AI服务展开研究，选取了以新闻资讯为主的云听平台、以综合性音频服务为主的喜马拉雅和以儿童教育内容为主的恐龙贝克平台中的AI传播内容进行具体研究，总结其应用功能和应用实践结果。本文在对典型案例进行分析的基础上，结合SPSS软件进行问卷分析，总结出AI语音合成技术在网络音频平台中的应用困局，即市场准入门槛低、AI应用功能缺位、合成语音质量难保证、存在侵权风险等。针对以上问题，本文给出强化有声市场管理、优化AI音频服务功能、提升合成语音质量、坚持伦理与法律原则的解决对策，希望通过本研究研究，可以为AI语音合成技术在网络音频领域的正面应用与发展提供参考，构建出的全新声音环境。

   With the continuous changes in media technology, the status of "sound" as an auxiliary medium is changing, and the value of sound and audio is gradually being explored. In the context of mobile information dissemination, online audio has begun to enter the public's perspective, and the flourishing of major audio platforms has brought new development opportunities to the audio industry. In recent years, online audio header platforms have joined the ranks of using AI speech synthesis technology, providing listeners with different auditory experiences through intelligent, user-friendly, and professional services. In the context of the auditory culture shift, AI speech synthesis technology has created a new artificial sound landscape, so it is necessary to explore the presentation of AI sound elements in online audio content.

   This study first traces the origin of AI speech synthesis technology and online audio platforms, and finds that deep learning is the mainstream method of speech synthesis at present. However, the application of AI speech synthesis technology in online audio platforms has entered a vigorous period of development. In fact, AI works have already existed on video platforms, and now, with the help of the development of online audio, they have significant development potential. This article conducts a research on representative online audio AI services in the current market, selecting AI voice works from cloud listening platforms that focus on news and information, Himalayas that focus on comprehensive audio services, and Dinosaur Baker platform that focuses on children's educational content, to conduct a specific study and summarize their application functions and practical results. Based on the analysis of typical cases and questionnaire analysis using SPSS software, this article summarizes the application difficulties of AI speech synthesis technology in network audio platforms, including low market entry barriers, lack of AI application functions, difficulty in ensuring the quality of synthesized speech, and risk of infringement. In response to the above issues, this article provides solutions to strengthen sound market management, optimize AI audio service functions, improve the quality of synthesized speech, and adhere to ethical and legal principles. It is hoped that through this research, it can provide reference for the positive application and development of AI voice synthesis technology in the field of network audio, creating a new sound environment.