ElevenLabs 发布 Eleven v4 语音模型，支持提示词设计音色与单文件克隆
Same guy, completely different voices 🔥 Try it out with @ElevenLabs ElevenLabs' new Eleven v4 model lets you design any voice from a text prompt, or clone one from a single audio file, then direct the delivery with tags for whispers, laughs and attitude. Pair it with Seedance 2.5 to drop those voices straight into your AI video, giving filmmakers, creators and advertisers full control over every character, accent and performance. Plus Eleven v4 generations are free until Oct 12 for Creator+ plans - a perfect time to start! #ElevenLabs #AIVoice #VoiceCloning #AIVideo #CreativeAI
ElevenLabs 推出新模型 Eleven v4，可从文本提示词设计任意音色，或从单个音频文件克隆声音，并用标签控制耳语、笑声和语气。该模型可与 Seedance 2.5 配合，把生成的语音直接放进 AI 视频；Creator+ 套餐的 Eleven v4 生成在 10 月 12 日前免费。
用同一段台词分别演示 Eleven v4 的提示词设计音色与单文件克隆，再接入 Seedance 2.5 生成视频，直观展示同一人多种声音的差异。
- 浏览
- 6.7万
- 点赞
- 3444
- 评论
- 106
- 分享
- 477
- 收藏
- 1888
约为这个来源平时的 1.6 倍
平台标注：AI 生成内容
数字来自平台，为最近一次抓取时的数据。
以下是视频口播的自动转写，可能有错字。
Voice is just got a massive upgrade in ai video. Same guy different voice. Now put the money in the bag. I think somebody did a murder.
We should call the police. We are the police. Oh yeah. So now you can play any character, any accent or any attitude babe.
You just have to describe the voice. And it'll do the subtle things the little things like the whispers and even the laughs ha ha ha here's how you can do it for your video to make these voices you can use the brand new 11 v 4 model which allows you to use voice design where you can prompt your own voices or you can use instant voice clone which allows you to upload an audio file and clone that voice if。
Using 11 v4, you can write in your prompt and use auto tax to direct the voice style. This is crazy. Now head to image and video and select c-dance 2.5 and upload your audio file in here.
Upload your video references and image references and write in your prompt. If you want to try it yourself, type ai in the comments and i'll send you a link.
来源：TikTok · Rourke Heath · tiktok.com