AI 模型和平台
Convoso 使无码声音 AI Agent 一般可用
在 Google 上的首选来源中添加 Unite.AI
Convoso 宣布于2026年11月3日表示其无码语音AI Agent已进入普遍可用状态，将自动领先资格和进站呼叫处理带入其出站销售平台。 该产品于2026年11月2日开始普遍提供。 它在将可供销售的前景转移给人类agents之前，先接触并限定线索。
agent原地建造成Convoso，使用与人类出行操作相同的运动工作流程，报告和合规控制。 它处理重复性的初步接触和资格，而入境职能包括回答例行问题、安排预约和向适当的小组或部门拨打电话。 发布因此新增是现有运动的自动层，AI活动与人手交接一起管理。
交谈工具和操作控制
Convoso报告了大约600毫秒的响应时间和2 000万次销售对话的培训，以及准备用于前后销售对话的中断处理。 该系统提供AI模型和文本转语音选项的选择，700多个声音，控制情感的直觉和速度，以及可选的语音克隆。 它旨在识别例行的确认而不无必要地停止，导航反对，维护上下文并在多语种通话时切换语言。
Auto Workflow可以使用历史调用录音来帮助创建资格工作流程，而不需要每条路径手动脚本。 一个浏览器测试功能允许用户在现场通话之前先与agent交谈。 如果主AI模型失败，一个提供者倒置链会自动将呼叫转移到备份模型。 agent在SIP/RTP上特别与Convoso Dialer平台集成，因此其目前的集成是Convoso自己的拨号堆。
运动和呼叫业绩报告既包括人工智能，也包括人类活动，从最初的外联到资格认证和转移。 现有的外联控制包括不拨号检查、州一级的拨号限制、每日时间规则以及根据每个客户的政策和风险承受能力进行的PII修改。
早期客户部署
Convoso 称 Tort Take 专业人员在最初的36小时测试中进行了150多次接触，其中约10-12%经过验证并被温暖地转移给人类专家。 团队计划将测试扩大到更先进的铅屏工作流程。
称组织每天管理数以万计的外出电话。 他描述agent的作用是进行初步连接，并在专家增加最大价值时让他们参与对话。
Ace Building Materials在单独的客户故事中提供了一个入境的例子。 75岁家庭拥有的供应商有5名前台销售人员，他们把注意力分散在当面的客户和接听的电话之间，让其他打电话者排队等待。 其命名为George的AI agent现在回答了关于商店时间和产品的常见问题，并在无法解决请求时将呼叫者与发送者或其他适当员工连接起来。
George认得西班牙语自动从英语切换连长Josh Zakarya 称的部署在Convoso的指导下花了几周时间。 他的新增产品和库存信息，以及过去客户通话的录音，以改善agent的反应。 在Ace,agent吸收了溢出电话，而销售人员则留在已经摆在他们面前的客户。
早期的账号显示agent服务于呼叫工作流程中的两点：在Tort Intake专业公司获得大量初始资格，在Ace提供超时客户服务。 在这两种情况下，人事交接都是中央专家得到证实的前景，而雇员则得到的呼叫者的要求需要重视，而不是常规的答案。 发布的实际意义是将这些交割、报告和控制放在同一个Convoso行动中。
本文由 AI图鉴自动抓取并翻译，版权归原作者所有；如需查看原文请点下方链接。
AI Models & Platforms
Convoso Makes No-Code Voice AI Agent Generally Available
Add Unite.AI to your preferred sources on Google
Convoso announced on November 3, 2026 that its no-code Voice AI Agent had entered general availability, bringing automated lead qualification and inbound call handling into its outbound sales platform. The product became generally available on November 2, 2026. It engages and qualifies leads before transferring sales-ready prospects to human agents.
Built natively into Convoso, the agent uses the same campaign workflows, reporting and compliance controls as the human outbound operation. It handles repetitive initial engagement and qualification, while inbound functions include answering routine questions, scheduling appointments and routing callers to the appropriate team or department. The launch therefore adds an automated layer to existing campaigns, with AI activity and human handoffs managed together.
Conversation Tools and Operating Controls
Convoso reports approximately 600-millisecond response times and training on 20 million sales conversations, alongside interruption handling intended for back-and-forth sales conversations. The system offers a choice of AI models and text-to-speech options, more than 700 voices, controls for emotional tonality and speed, and optional voice cloning. It is designed to recognize routine acknowledgements without stopping unnecessarily, navigate objections, maintain context and switch languages during multilingual calls.
Auto Workflow can use historical call recordings to help create qualification workflows without requiring every path to be scripted manually. A browser test feature lets users speak with an agent before deploying it on live calls. If the primary AI model fails, a provider fallback chain automatically shifts calls to a backup model. The agent integrates specifically with the Convoso Dialer Platform over SIP/RTP, so its current integration is with Convoso’s own dialer stack.
Campaign and call-performance reporting covers both AI and human activity, from initial outreach through qualification and transfer. Available outreach controls include do-not-call checks, state-level dialing restrictions, time-of-day rules and PII redaction, configured according to each customer’s policies and risk tolerance.
Early Customer Deployments
Convoso said Tort Intake Professionals made more than 150 contacts during an initial 36-hour test, with approximately 10–12% validated and warm-transferred to human specialists. The team planned to expand testing into more advanced lead-screening workflows.
Nicholas Ward, Tort Intake Professionals’ director of business operations, said the organization manages tens of thousands of outbound calls daily. He described the agent’s role as making initial connections and bringing specialists into conversations when they add the most value.
Ace Building Materials provides an inbound example in a separate customer story. The 75-year-old family-owned supplier has five front-counter salespeople who were splitting their attention between in-person customers and incoming calls, leaving other callers waiting in a queue. Its AI agent, named George, now answers common questions about store hours and products and connects callers with dispatch or another appropriate employee when it cannot resolve a request.
George recognizes Spanish and switches from English automatically. Company president Josh Zakarya said deployment took a couple of weeks with Convoso’s guidance. He adds product and inventory information, along with recordings of past customer calls, to improve the agent’s responses. At Ace, the agent absorbs overflow calls while salespeople stay with the customers already in front of them.
The early accounts show the agent serving two points in the call workflow: high-volume initial qualification at Tort Intake Professionals and overflow customer service at Ace. In both, the human handoff is central—specialists receive validated prospects, while employees receive callers whose requests need attention beyond routine answers. The practical significance of the launch is the placement of those handoffs, reporting and controls inside the same Convoso operation.
Auto-fetched and translated by AI Atlas. All rights belong to the original author.