|
听力频道 > 英语视频 > 正文 |
类人机器人阿美卡又变强了!多种语言切换自如,不过说中文时犯了个小错…… 当地时间4月6日,英国“工程艺术(Engineered Arts)”公司发布了类人机器人阿美卡(Ameca)的最新演示视频。视频中,阿美卡按要求用日语、德语、法语、中文回答问题,多国语言切换自如,还能分英音和美音。↓ (视频戳这里)
“This Ameca demo uses GPT3 for conversation and translation. DeepL is used for language detection. The voices are Amazon Polly Neural voices.” 视频中,研发人员问阿美卡,是否会多种不同语言。阿美卡用默认的英式英语回答说,“我会说很多种语言,包括德语、英语、法语、日语、中文等等。” “I can speak many languages, including German, English, French, Japanese, Chinese, and much more.”
“它说的法语带有明显的魁北克(加拿大法语)口音。法国法语、加拿大法语、瑞士法语和比利时法语(可能我还忽略了其他地方的法语),乍听起来很相似,但都有自己的发音/口音,在词汇和其他文化特色上也有差异。也许这只是亚马逊在法语中加载了错误的‘口音’,但我得指出来让你们知道。” “The French voice has a noticeable Quebec (Canadian French) accent. ‘French’ French, Canadian French, Swiss French and Belgian French (and I may forget other Frenches), can look like similar at first but they do have their own pronunciation/accent, and they have differences in vocabulary and other cultural specialties. Perhaps this is just Amazon loading the wrong "accent" to the French language but I had to inform you about that just so you know. ”
“We are working on a demo using ElevenLabs voice cloning, it adds some complexity because of the additional phoneme and viseme generation for lip sync. All of this is put together on our Tritium software platform, we will be releasing a beta public version of in the coming months. It includes virtual Ameca and support for importing other robot models in SDF format.”
|
|
主办
|
中报二十一世纪(北京)传媒科技有限公司版权所有,未经书面授权,禁止转载或建立镜像。 主办单位:中国日报社 Copyright by 21st Century English Education Media All Rights Reserved 版权所有 复制必究 网站信息网络传播视听节目许可证0108263 京ICP备2024066071号-1 京公网安备 11010502033664号 |