xAI 发布 Grok Voice 3.0 语音模型

来源:xAI:News(网页) 2026年10月14日 08:00 AIHOT 评分:66 精选
摘要:xAI 发布 speech-to-speech 模型 Grok Voice 3.0,在 Artificial Analysis 的 Speech-to-Speech Index 以 86.9 分排名第一,时延低于一秒。
推荐理由:官方发布给出多项第三方基准分数、时延和 API 定价,读者可以据此评估它在语音智能体场景的性价比。
# xAI 发布 Grok Voice 3.0 语音模型 > 原标题:Oct 14, 2026 Introducing Grok Voice 3.0 The world’s most intelligent voice model. Read More - 来源:xAI:News - 发布时间:2026-10-14T00:00:00.000Z - AIHOT:https://aihot.news/items/qld0z3p2gztpolbmviizsm66u - 原文:https://x.ai/news/grok-voice-3 ## 摘要 xAI 发布 speech-to-speech 模型 Grok Voice 3.0,在 Artificial Analysis 的 Speech-to-Speech Index 以 86.9 分排名第一,时延低于一秒。 ## 推荐理由 官方发布给出多项第三方基准分数、时延和 API 定价,读者可以据此评估它在语音智能体场景的性价比。 ## 正文 · 中文译文 今天,我们宣布推出 Grok Voice 3.0,全球最智能的语音到语音模型。它在解决客户支持通话方面创下新纪录,延迟处于同类领先水平,低于一秒,且成本低于其他前沿模型。Grok Voice 每天为 Tesla 和 SpaceX 的数百万次对话提供支持,今天已在 API 中上线,价格与前代相同。 ## [最准确的语音模型](https://x.ai/news/grok-voice-3#the-most-accurate-voice-model) 在 Artificial Analysis 的独立测试中,Grok Voice 3.0 在 Speech-to-Speech Index 上排名第一,得分为 86.9。该指数综合了语音推理、τ\-Voice,以及盲测一对一通话中的偏好和任务成功率。Grok Voice 3.0 在 Full Duplex Bench 上也获得 97% 的分数,该基准衡量其处理停顿、轮流发言和打断的自然程度。 Grok Voice 3.0 与竞争模型在语音到语音基准上的对比,依据 Artificial Analysis。 | 基准 | Grok Voice 3.0(本次发布) | Grok Voice 2.0 | GPT-Live-1(Astra,Medium) | Gemini 3.8 Live(Ext. Thinking,High) | | --- | --- | --- | --- | --- | | 总体
AA Speech-to-Speech Index | 86.9 | 81.3 | 83.2 | 82.6 | | 语音推理
Big Bench Audio | 98.5% | 97.2% | 90.1% | 97.7% | | 人类偏好
Speech Agent Arena Elo | 1070 | 1011 | 1048 | 990 | | 任务完成
任务成功率 | 94.8% | 94.6% | 87.4% | 89.1% | | 对话动态
Full Duplex Bench | 97.0% | 95.1% | 94.9% | 91.9% | τ\-Voice 衡量智能体处理带有不同口音、背景噪音和打断的支持通话的能力。Grok Voice 3.0 得分为 71.6%,与 GPT-Live-1(Astra)相当,但延迟低 40%,在智能体智能与响应速度之间树立了新的前沿。 它的运行成本也低于同类模型。Artificial Analysis 测量了每个模型在其 Big Bench Audio 运行中完成任务所需的成本,Grok Voice 3.0 低于 GPT-Live-1(Astra,Medium)和 GPT-Realtime-2.1。 每任务成本低于同类模型 Artificial Analysis 的 Big Bench Audio 运行中每任务平均成本,美元。越低越好。 在 [EVA-Bench](https://servicenow.github.io/eva) 上,即 ServiceNow 面向企业语音智能体的基准,Grok Voice 3.0 取得了 74% 的最高准确率得分,并且在准确率和客户体验的综合评分中表现最佳。 > “Grok Voice 3.0 在 EVA-Bench 上树立了新的帕累托前沿。EVA-Bench 是 ServiceNow 的开源基准,在 200 多个真实企业场景中对语音智能体的准确率和体验进行评分。它将顶尖准确率与自然、响应迅速的对话相结合,为企业在前线客户服务中对 AI 的期待树立了新标准。” Nitin Aggarwal,ServiceNow 产品管理高级总监,负责 Agentic Evals 和 RAI ## [被 Tesla 和 SpaceX 数百万客户使用](https://x.ai/news/grok-voice-3#used-by-millions-of-customers-at-tesla--spacex) Grok Voice 3.0 每天在 Tesla 车辆中处理数百万次对话。驾驶员可以使用 Connectors 处理简单请求,例如查看日历或回复电子邮件,并使用 Grok Bot 处理更大的任务,例如点咖啡或预订晚餐——全程完全免手操作([在 X 上查看](https://x.com/Tesla/status/2102430656349544590))。Grok Voice 3.0 也为 Grok Bot 的语音模式提供支持。 Starlink 呼叫中心使用 Grok Voice 每天以多种语言接听 20,000 通客户来电。约 70% 无需人工介入即可解决,四分之一的销售通话在来电者挂断前完成购买。一个基于 Grok Voice 3.0 构建的单一代理即可验证客户身份、排查硬件故障、发放信用额度并寄送替换设备。 ## [定价与可用性](https://x.ai/news/grok-voice-3#pricing-and-availability) Grok Voice 3.0 即日起在 Voice Agent API 中提供,价格为 [$0.08 / 分钟($4.80 / 小时)](https://docs.x.ai/developers/pricing?campaign=voice-3-blog#voice-api-pricing),与 Grok Voice 2.0 相同。使用 `grok-voice-latest` 的代理已自动切换,无需任何更改。 ## 开始使用 Grok Voice 构建 [免费试用](https://console.x.ai/team/default/voice/agents?campaign=voice-3-blog&utm_source=website&utm_medium=referral&utm_campaign=voice-3-blog)[查看文档](https://docs.x.ai/developers/model-capabilities/audio/voice?campaign=voice-3-blog) ## 正文 · 原文 Today, we’re announcing Grok Voice 3.0, the world’s most intelligent speech-to-speech model. It sets a new record for resolving customer support calls, has class-leading latency at under one second, and costs less than other frontier models. Grok Voice powers millions of conversations a day across Tesla and SpaceX, and is available today in the API at the same price as its predecessor. ## [The Most Accurate Voice Model](https://x.ai/news/grok-voice-3#the-most-accurate-voice-model) In Artificial Analysis’s independent testing, Grok Voice 3.0 ranks #1 on the Speech-to-Speech Index, with a score of 86.9. The Index combines speech reasoning, τ\-Voice, and preference and task success from blind head-to-head calls. Grok Voice 3.0 also scores 97% on Full Duplex Bench, a measure of how naturally it handles pauses, turn-taking, and interruptions. Speech-to-speech benchmark comparison between Grok Voice 3.0 and competing models, per Artificial Analysis. | Benchmark | Grok Voice 3.0 (this release) | Grok Voice 2.0 | GPT-Live-1(Astra, Medium) | Gemini 3.8 Live(Ext. Thinking, High) | | --- | --- | --- | --- | --- | | Overall
AA Speech-to-Speech Index | 86.9 | 81.3 | 83.2 | 82.6 | | Speech Reasoning
Big Bench Audio | 98.5% | 97.2% | 90.1% | 97.7% | | Human Preference
Speech Agent Arena Elo | 1070 | 1011 | 1048 | 990 | | Task Completion
Task Success Rate | 94.8% | 94.6% | 87.4% | 89.1% | | Conversational Dynamics
Full Duplex Bench | 97.0% | 95.1% | 94.9% | 91.9% | τ\-Voice measures how well agents handle support calls with diverse accents, background noise, and interruptions. Grok Voice 3.0 scores 71.6%, comparable with GPT-Live-1 (Astra), but at 40% lower latency, setting a new frontier in agent intelligence versus responsiveness. It's also cheaper to run than comparable models. Artificial Analysis measures what each model costs to complete a task in its Big Bench Audio run, and Grok Voice 3.0 comes in under both GPT-Live-1 (Astra, Medium) and GPT-Realtime-2.1. Lower cost per task than comparable models Average cost per task on Artificial Analysis's Big Bench Audio run, USD. Lower is better. On [EVA-Bench](https://servicenow.github.io/eva), ServiceNow’s benchmark for enterprise voice agents, Grok Voice 3.0 achieves the highest accuracy score at 74% and is the best-performing model when jointly scored across accuracy and customer experience. > “Grok Voice 3.0 sets a new Pareto frontier on EVA-Bench, ServiceNow’s open-source benchmark scoring voice agents on accuracy and experience across 200+ realistic enterprise scenarios. It pairs top accuracy with a natural, responsive conversation, setting a new standard for what enterprises can expect from AI on the front line of customer service.” Nitin Aggarwal, Senior Director of Product Management, Agentic Evals and RAI, ServiceNow ## [Used by Millions of Customers at Tesla & SpaceX](https://x.ai/news/grok-voice-3#used-by-millions-of-customers-at-tesla--spacex) Grok Voice 3.0 handles millions of conversations a day in Tesla vehicles. Drivers can use Connectors for simple requests like checking their calendar or answering an email, and Grok Bot for bigger tasks like ordering a coffee or booking dinner—all completely hands-free ([see it on X](https://x.com/Tesla/status/2102430656349544590)). Grok Voice 3.0 also powers Grok Bot’s voice mode. Starlink call centers use Grok Voice to answer 20,000 customer calls every day across multiple languages. Roughly 70% are resolved with no human in the loop, and one in four sales calls ends in a purchase before the caller hangs up. A single agent built on Grok Voice 3.0 verifies customers, troubleshoots hardware, issues credits, and ships replacement equipment. ## [Pricing and Availability](https://x.ai/news/grok-voice-3#pricing-and-availability) Grok Voice 3.0 is available today in the Voice Agent API at [$0.08 / min ($4.80 / hr)](https://docs.x.ai/developers/pricing?campaign=voice-3-blog#voice-api-pricing), the same price as Grok Voice 2.0. Agents using `grok-voice-latest` have already been switched over automatically, with no changes needed. ## Start building with Grok Voice [Try It Free](https://console.x.ai/team/default/voice/agents?campaign=voice-3-blog&utm_source=website&utm_medium=referral&utm_campaign=voice-3-blog)[View Docs](https://docs.x.ai/developers/model-capabilities/audio/voice?campaign=voice-3-blog)