“语音身份”与“语音秩序”:ElevenLabs引发的数字声音景观重构及其风险  

Speech Identity and Speech Order:ElevenLabs,Its Reconstruction of the Digital Soundscape,and Its Risks

作　　者：韩晓晔[1] HAN Xiaoye(Film-Television and Communication College,Shanghai Normal University,Shanghai 200234)

机构地区：[1]上海师范大学影视传媒学院,上海200234

出　　处：《文化艺术研究》2025年第1期48-56,113,共10页Studies in Culture and Art

摘　　要：人工智能技术的突飞猛进正在重塑人类感知和理解世界的方式。继Midjourney、Stable Diffusion等引领视觉生成式AI掀起新浪潮后,以ElevenLabs为代表的先进语音合成平台也标志着音频生成式AI的崛起。ElevenLabs凭借其高度拟真、富有情感、支持个性化定制的AI合成语音,造就了一种前所未有的声音景观,但也带来了人声合成泛滥对“语音身份”认同的解构与冲击,由此造成社会传统“语音秩序”的崩溃和重构,带来语音身份多变与身份识别困境、语音身份盗用与社会信任危机、数字语音规训与意识形态操控、虚假语音煽动与社会冲突威胁等风险。AI is reshaping how humans perceive and understand the world.Following the visual revolution led by Midjourney and Stable Diffusion,ElevenLabs represents the rise of audio-generative AI.Known for its highly realistic,emotionally expressive,and customizable synthetic speech,ElevenLabs has created an unprecedented soundscape.However,it also has led to the misuse of synthetic speech,deconstructing speech identity,and causing the collapse and reconstruction of the traditional speech order.This shift introduces risks including the instability of speech identity,challenges in identity recognition,speech identity theft and a crisis of social trust,digital speech discipline and ideological manipulation,as well as fake speech and social conflict.

关 键 词：ElevenLabs 语音身份 语音秩序 数字人 人机共生 技术向善 

分 类 号：G206.3[文化科学—传播学]