Gradium发布Voice Design：通过文本提示秒级生成合成语音
Gradium Launches Voice Design: Write a Prompt, Get a Brand New Synthetic Voice in Seconds
Prompt-to-Voice技术落地且完全免费，解决了长尾口音定制难题，做语音Agent的同学值得直接接入测试其效果。
Voice agent teams keep hitting the same wall. The catalog holds 400 voices and the brief asks for the one that is not in it: a Quebecoise receptionist for a Montreal dealership, a narrator in his sixties with lecture hall authority. Briefs outnumber any catalog, and cloning closes the gap one speaker at a time, each carrying sourcing, consent and a licence.
语音代理团队总是撞上一堵墙。目录中收录了 400 种声音，而需求简报要求的却是其中没有的一种：一位蒙特利尔经销商的魁北克口音接待员，或是一位拥有讲座厅权威感的六十多岁旁白。需求简报的数量远超目录，克隆技术逐一缩小差距，每次生成都涉及来源、同意和许可。
Gradium, the Paris-based voice AI company spun out of the Kyutai research lab, has shipped a different answer. Voice Design reads a written description and returns complete new voices in a few seconds. No reference audio, no speaker, no rights to clear.
总部位于巴黎、源自 Kyutai 研究实验室的语音 AI 公司 Gradium 给出了不同的答案。Voice Design 读取书面描述，并在几秒内返回完整的新声音。无需参考音频，无需特定说话人，也无需清理权利。
Is it deployable? Yes, Voice Design is live in the Gradium API and in Studio, free on every plan including the free tier, and a kept voice runs on the same streaming Text-to-Speech endpoint as any catalog voice, at the same latency and output formats.
它是否可部署？是的，Voice Design 已在 Gradium API 和 Studio 中上线，在所有计划（包括免费层级）上免费使用，且保留的声音与目录中的任何声音一样，在相同的延迟和输出格式下运行于相同的流式文本转语音端点。
The casting brief is the API
选角简报即为 API
The description is the only input the model gets. Gradium’s documentation lists the attributes it responds to, and they read like a casting call: gender, age band, accent or origin, pitch, pace, energy, timbre and resonance, register and manner, and the job the voice is doing. Descriptions run 1 to 500 characters in English, French, Spanish, Portuguese or German. Gradium advises ending with the intended use, because it steers delivery and register rather than only the colour of the voice.
描述是模型获得的唯一输入。Gradium 的文档列出了它所响应的属性，读起来就像选角通知：性别、年龄段、口音或起源、音高、语速、能量、音色和共鸣、音域和风格，以及声音所承担的任务。描述长度为 1 到 500 个字符，支持英语、法语、西班牙语、葡萄牙语或德语。Gradium 建议以预期用途结尾，因为它能引导交付方式和音域，而不仅仅是声音的色彩。
One request returns 1 to 5 candidates, typically ready in 3 to 5 seconds. They are variations on a single character, so a different character means a different description, not more samples.
一次请求返回 1 到 5 个候选者，通常在 3 到 5 秒内就绪。它们是单一角色的变体，因此不同的角色意味着不同的描述，而非更多的样本。
From candidate to production voice
从候选者到生产用声音
The flow is four calls. POST /voice-generator/generate mints candidate ids with ready: false. GET /voice-generator/embeddings polls until they flip. Each candidate auditions through the ordinary TTS endpoint, using the candidate id as voice_id. POST /voices/from-embedding promotes the one you keep.
流程包含四次调用。POST /voice-generator/generate 生成 ready: false 的候选者 ID。GET /voice-generator/embeddings 轮询直到状态变为 true。每个候选者通过普通的 TTS 端点进行试音，使用候选者 ID 作为 voice_id。POST /voices/from-embedding 将你保留的那个提升为正式声音。
Candidates carry three restrictions converted voices do not: audition text is capped at 100 characters, they are REST only, and the TTS WebSocket and Speech-to-Speech reject them. Unconverted candidates are deleted after 30 days. Converting is free, clears the expiry, and uses one custom voice slot shared with clones. The free tier holds 5, paid plans 1,000.
候选者携带三种转换后的声音所不具备的限制：试音文本限制在 100 个字符以内，仅支持 REST，TTS WebSocket 和语音转语音功能会拒绝它们。未转换的候选者在 30 天后删除。转换是免费的，清除过期时间，并占用一个与克隆共享的自定义声音槽位。免费层级持有 5 个，付费计划持有 1,000 个。
Sampling is deliberately non-deterministic. Gradium team expands the description first, and that expansion varies per request, so the same prompt with a fixed seed still yields a different voice.
采样故意是非确定性的。Gradium 团队首先扩展描述，且该扩展因请求而异，因此即使使用固定种子，相同的提示词仍会产生不同的声音。
The benchmark, and how to read it
基准测试及如何解读它
Gradium ran a blind pairwise listening test on accent prompts across six voice design systems reachable through public APIs and five languages. Native speakers heard two unlabelled clips and picked the closer match, or a tie. Across 7,627 comparisons, Gradium reports a 72.6% win rate against the field, 13.6 points ahead of ElevenLabs eleven_ttv_v3 at 59.0%, followed by Inworld at 44.8%, Fish Audio at 36.7% and MiniMax at 31.7%. Win rate is wins plus half of ties, so 50% is par. Gradium placed first in all five languages. The widest margins came on regional accents that most catalogs flatten: Quebecois French at 97%, Rioplatense Spanish at 86%, Bavarian German at 85%, Colombian Spanish and African Portuguese at 83%.
Gradium对可通过公共API访问的六个语音设计系统以及五种语言下的口音提示进行了盲听配对测试。母语使用者会听到两个未标记的音频片段，并选择更接近匹配的一个，或选择平局。在7,627次比较中，Gradium报告其胜率为72.6%，领先于ElevenLabs eleven_ttv_v3（59.0%）13.6个百分点，随后是Inworld（44.8%）、Fish Audio（36.7%）和MiniMax（31.7%）。胜率计算方式为胜场数加上平局数的一半，因此50%为基准水平。Gradium在所有五种语言中均排名第一。优势最明显的领域在于大多数目录会抹平的地区口音：魁北克法语达97%，拉普拉塔西班牙语达86%，巴伐利亚德语达85%，哥伦比亚西班牙语和非洲葡萄牙语均为83%。
A model judge over the same prompt set agreed. Gemini 3.1 Pro rated single unlabelled clips from 1 to 5 and produced the same ranking: Gradium 4.06, ElevenLabs 3.86, Inworld 3.64, Fish Audio 3.51. Separately, the product page claims 83.4% prompt adherence on the English split of InstructTTSEval, the academic benchmark for instruction following in TTS. (Note: All of these numbers are vendor designed and vendor run.)
同一提示集上的模型评审也得出了相同结论。Gemini 3.1 Pro对单个未标记片段进行1至5分的评分，并产生了相同的排名：Gradium为4.06分，ElevenLabs为3.86分，Inworld为3.64分，Fish Audio为3.51分。此外，产品页面声称在InstructTTSEval的英文拆分部分（TTS指令遵循的学术基准）上，提示遵循率达到83.4%。（注：所有这些数字均由供应商设计和运行。）
Key Takeaways
关键要点
- Voice Design turns a 500 character description into up to 5 new voices in seconds, no reference audio required.
- It is live and free on every Gradium plan, in the API and Studio, across 5 languages.
- Vendor run blind tests put it at a 72.6% win rate over 7,627 comparisons, first in all 5 languages.
- Kept voices become a normal voice_id on REST, WebSocket and Speech-to-Speech.
- Sampling is non-deterministic, so an unsaved candidate is gone for good after 30 days.
- Voice Design能在几秒钟内将500个字符的描述转化为多达5个新声音，无需参考音频。
- 它在所有Gradium计划、API和Studio中均已上线且免费使用，支持5种语言。
- 由供应商运行的盲测显示，在7,627次比较中，其胜率为72.6%，在所有5种语言中均排名第一。
- 保存后的声音在REST、WebSocket和语音到语音（Speech-to-Speech）接口中将变为常规的voice_id。
- 采样过程是非确定性的，因此未保存的候选声音在30天后将永久消失。
Check out the Technical details and Docs. Also, feel free to follow us on Twitter and don’t forget to join our 150k+ML SubReddit and Subscribe to our Newsletter. Wait! are you on telegram? now you can join us on telegram as well.
查看技术细节和文档。此外，欢迎在Twitter上关注我们，别忘了加入我们有15万+成员的ML SubReddit，并订阅我们的新闻通讯。等等！你在Telegram上吗？现在你也可以加入我们的Telegram群组。
Need to partner with us for promoting your GitHub Repo OR Hugging Face Page OR Product Release OR Webinar etc.? Connect with us
需要与我们合作推广你的GitHub仓库、Hugging Face页面、产品发布或网络研讨会等？请联系我们
The post Gradium Launches Voice Design: Write a Prompt, Get a Brand New Synthetic Voice in Seconds appeared first on MarkTechPost.
本文《Gradium推出Voice Design：输入提示，秒获全新合成声音》首发于MarkTechPost。
更进一步：量化金融体系
看懂新闻只是起点——沿量化金融路径，把它变成能交付的工程能力