<p>，/PRNewswire/ -- 随着人工智能深入融入人类生活并重塑连接性，提升社交场景中互动体验所需的基础能力是什么？</p>
<p>最近，Soul App 升级了其自研的端到端全双工语音通话模型。重新定义 “全双工”，新模型摒弃了传统概念，如 “VAD（语音活动检测，通常用于检测语音开始/结束）” 和 “延迟”，打破了行业标准的 “轮流” 互动模式。</p>
<p>继续阅读</p>
<p><img src="https://imageproxy.pbkrs.com/https://www.prnewswire.com/news-releases/null?x-oss-process=image/auto-orient,1/interlace,1/resize,w_1440,h_1440/quality,q_95/format,jpg" alt="" original-src="https://imageproxy.pbkrs.com/https://www.prnewswire.com/news-releases/null"/></p>
<p>相反，它赋予人工智能自主决定发言时机的能力，例如主动打破沉默、适当地打断用户、在说话时倾听、感知时间语义、实现并行讨论等。该模型还支持多维感知（包括时间、环境和事件意识）和自然语音特征（例如，填充词、口吃、明显的情感波动），使人工智能的互动更加 “人性化”，提供沉浸式、逼真的语音体验。</p>
<p>升级后的全双工模型将很快在 Soul 进行 beta 测试，随后将在一对一的互动场景中部署，如数字人通话和人工智能配对。Soul 的团队还在探索其在群体场景中的应用，使人工智能能够在适当时机参与对话，扩展话题，促进多样化的关系网络。</p>
<p>Soul App 的首席技术官陶铭表示：“社交互动是情感和信息价值的交换。Soul 始终致力于利用创新技术和产品解决方案，提供更智能、更沉浸、更高质量的互动体验，让孤独感消失。”</p>
<p><strong>全双工语音通话：重新定义人工智能社交互动</strong></p>
<p>由于技术限制，人机对话往往遵循 “问答” 格式（用户提问 → 人工智能回答），其中延迟或打断会破坏沉浸感。</p>
<p>在 2024 年，Soul 推出了其自研的端到端全双工语音模型，具有超低延迟、快速自动打断、超真实的语音表达和情感感知。它可以直接解读复杂的听觉输入，并支持高度拟人化的多语言风格。为了进一步实现日常生活般的对话和 “人性化” 的陪伴，Soul 现在升级了模型，具备以下能力：</p>
<p><strong><em>1.</em></strong> <strong><em>全双工互动：人工智能获得自主决策能力</em></strong></p>
<p>新模型实现了响应的流预测、倾听和打断。人工智能自主决定何时发言，实现真正的端到端全双工互动，人工智能和用户可以同时交谈（例如，辩论、争论、唱歌），适当地打断彼此，或主动打破沉默以引入话题。</p>
<p>这种自主性使人工智能能够掌握互动时机，显著增强对话的自然性，并在延续的多轮对话中实现沉浸式、现实生活般的交流。</p>
<p><strong><em>2.</em></strong> <strong><em>口语化与情感表达：更生动的互动</em></strong></p>
<p>为了让人工智能的互动更具人性化，该模型在多个维度上实现了全面增强，包括情感表达、声音特征和对话内容，使其更接近现实世界的沟通。</p>
<p>例如，在情感表达方面，除了基础能力如笑声、哭泣或愤怒外，升级后的模型提供了更明显的声音波动，随着对话自然演变。其发音现在融入了有机的语音元素，如填充词、偶尔的口吃、常见的口头禅、咳嗽和其他日常语音细微差别。此外，人工智能生成的对话倾向于使用口语化和社会流动的语言，而不是僵硬的书面语言模式。</p>
<p><strong><em>3.</em></strong> <strong><em>情境意识：时间、事件、环境</em></strong></p>
<p>基于纯自回归架构与统一文本/音频生成（统一模型），该模型利用强大的 LLM 能力将个性、时间、环境和情境对话整合到人工智能的响应中。这使得具有感知能力和理解能力的人工智能能够更好地塑造 “数字个性”，创造丰富的故事情节，并将互动转变为真正的 “情感和信息的交流”。</p>
<p>值得注意的是，Soul 的人工智能团队目前正在探索如何将其全双工语音通话模型扩展到多人场景。例如，在群体语音对话中，人工智能利用其自主决策能力识别最佳发言时机，有效促进话题讨论和扩展，并作为积极参与者无缝融入真实的社交动态。</p>
<p><strong>将人工智能融入社交网络：提供情感和信息价值</strong></p>
<p>基于对社交动态的深刻洞察，Soul 快速在应用层实施技术，并根据用户反馈优化产品。其人工智能路线图专注于两条路径：“人工智能帮助用户交朋友（人工智能辅助社交）” 和 “人工智能与用户成为朋友（人机互动）”。相应的功能，如情感智能的 “人工智能伴侣” 和聊天助手 “人工智能聊天助手”，已获得用户的热烈采用。</p>
<p>人机互动特别旨在通过类人能力提供情感和信息价值。根据 Soul 的 Just So Soul Institute（2025 年 Z 世代人工智能使用报告，2025 年 3 月；样本：3680）：</p>
<ul>
<li>几乎 40% 的年轻人每天使用人工智能进行情感陪伴。</li>
<li>71.1% 愿意与人工智能交朋友（相比于 2024 年 Z 世代 AIGC 态度报告中的 32.8%）。</li>
</ul>
<p>随着人工智能重塑 Z 世代对社交关系的认知，它也带来了新的需求。Soul 对 “人工智能伴侣” 的活跃用户进行的调查显示，约 60% 的人希望人工智能表现得更像人类。</p>
<p>这一全双工模型的升级显著提升了人工智能的互动能力，为人机交流注入了存在感和情感温度，将推动人工智能社交进入一个新纪元。</p>
<p>作为一个根植于真实人际关系的平台，Soul 致力于构建一个&#34;AI 存在&#34;与&#34;人类&#34;共存的社交生态系统。通过持续投资于前沿技术，Soul 使人工智能能够丰富用户的情感支持系统，丰富情感体验，最终提升个人的幸福感和归属感。</p>
<p>来源 Soul App</p>
<p><img src="https://imageproxy.pbkrs.com/https://rt.prnewswire.com/rt.gif/query-RGF0ZUlkPTIwMjUwNzEwJk5ld3NJdGVtSWQ9Q04yNzc2MSZUcmFuc21pc3Npb25fSWQ9MjAyNTA3MTAwMTIzUFJfTkVXU19VU1BSX19fX19DTjI3NzYx?x-oss-process=image/auto-orient,1/interlace,1/resize,w_1440,h_1440/quality,q_95/format,jpg" alt="" original-src="https://imageproxy.pbkrs.com/https://rt.prnewswire.com/rt.gif/query-RGF0ZUlkPTIwMjUwNzEwJk5ld3NJdGVtSWQ9Q04yNzc2MSZUcmFuc21pc3Npb25fSWQ9MjAyNTA3MTAwMTIzUFJfTkVXU19VU1BSX19fX19DTjI3NzYx"/></p>

22105

<p>Soul App 推出了一种升级的全双工语音通话模型，允许 AI 自主控制发言时机，从而增强互动体验。该模型打破了传统的语音检测方法，实现了同时对话和自然打断。它融入了情感表达和上下文意识，使 AI 互动更加人性化。该模型将很快进入测试阶段，旨在改善一对一和群体互动，促进更深层次的社交联系。Soul 的首席技术官强调了情感和信息交流在社交互动中的重要性，反映出年轻用户对 AI 陪伴日益增长的需求</p>

<p>, /PRNewswire/ -- As AI deeply integrates into human life and reshapes connectivity, what foundational capabilities are needed to enhance interactive experiences in social scenarios?</p>
<div class="lb-trans"><p>，/PRNewswire/ -- 随着人工智能深入融入人类生活并重塑连接性，提升社交场景中互动体验所需的基础能力是什么？</p>
</div><p>Recently, Soul App has upgraded its self-developed, end-to-end full-duplex voice call model. Redefining &#34;full-duplex&#34;, the new model abandons traditional concepts like &#34;VAD (Voice Activity Detection, commonly used to detect speech start/end)&#34; and &#34;latency&#34;, breaking away from the industry-standard &#34;turn-by-turn&#34; interaction pattern.</p>
<div class="lb-trans"><p>最近，Soul App 升级了其自研的端到端全双工语音通话模型。重新定义 “全双工”，新模型摒弃了传统概念，如 “VAD（语音活动检测，通常用于检测语音开始/结束）” 和 “延迟”，打破了行业标准的 “轮流” 互动模式。</p>
</div><p>Continue Reading</p>
<div class="lb-trans"><p>继续阅读</p>
</div><p><img src="https://imageproxy.pbkrs.com/https://www.prnewswire.com/news-releases/null" alt="" original-src="https://www.prnewswire.com/news-releases/null"/></p>
<p>Instead, it empowers AI to autonomously decide speaking timing, such as proactively breaking silence, appropriately interrupting users, listening while speaking, perceiving time semantics, enabling parallel discussions, and more. The model also supports multi-dimensional perception (including time, environment, and event awareness) and natural speech features (e.g., fillers, stammering, noticeable emotional fluctuations), making AI interactions more &#34;human-like&#34; and delivering an immersive, lifelike voice experience.</p>
<div class="lb-trans"><p>相反，它赋予人工智能自主决定发言时机的能力，例如主动打破沉默、适当地打断用户、在说话时倾听、感知时间语义、实现并行讨论等。该模型还支持多维感知（包括时间、环境和事件意识）和自然语音特征（例如，填充词、口吃、明显的情感波动），使人工智能的互动更加 “人性化”，提供沉浸式、逼真的语音体验。</p>
</div><p>The upgraded full-duplex model will soon enter beta testing on Soul and will later be deployed in one-on-one interactive scenarios like digital human calls and AI matchmaking. Soul's team is also exploring its application in group settings, enabling AI to join conversations at the right moment, extend topics, and foster diverse relationship networks.</p>
<div class="lb-trans"><p>升级后的全双工模型将很快在 Soul 进行 beta 测试，随后将在一对一的互动场景中部署，如数字人通话和人工智能配对。Soul 的团队还在探索其在群体场景中的应用，使人工智能能够在适当时机参与对话，扩展话题，促进多样化的关系网络。</p>
</div><p>Tao Ming, CTO of Soul App, stated, &#34;Social interaction is an exchange of emotional and informational value. Soul remains committed to leveraging innovative technology and product solutions to deliver smarter, more immersive, and higher-quality interactive experiences, making loneliness go away for all.&#34;</p>
<div class="lb-trans"><p>Soul App 的首席技术官陶铭表示：“社交互动是情感和信息价值的交换。Soul 始终致力于利用创新技术和产品解决方案，提供更智能、更沉浸、更高质量的互动体验，让孤独感消失。”</p>
</div><p><strong>Full-Duplex Voice Call: Redefining AI Social Interaction</strong></p>
<div class="lb-trans"><p><strong>全双工语音通话：重新定义人工智能社交互动</strong></p>
</div><p>Previously constrained by technical limitations, human-AI dialogues often followed a &#34;Q&amp;A&#34; format (user asks → AI responds), where latency or interruptions disrupted immersion.</p>
<div class="lb-trans"><p>由于技术限制，人机对话往往遵循 “问答” 格式（用户提问 → 人工智能回答），其中延迟或打断会破坏沉浸感。</p>
</div><p>In 2024, Soul launched its self-developed, end-to-end full-duplex voice model, featuring ultra-low latency, rapid auto-interruption, hyper-realistic voice expression, and emotional perception. It could directly interpret complex auditory inputs and support highly anthropomorphic multilingual styles. To further achieve daily-life-like conversations and &#34;human-like&#34; companionship, Soul has now upgraded the model with the following capabilities:</p>
<div class="lb-trans"><p>在 2024 年，Soul 推出了其自研的端到端全双工语音模型，具有超低延迟、快速自动打断、超真实的语音表达和情感感知。它可以直接解读复杂的听觉输入，并支持高度拟人化的多语言风格。为了进一步实现日常生活般的对话和 “人性化” 的陪伴，Soul 现在升级了模型，具备以下能力：</p>
</div><p><strong><em>1.</em></strong> <strong><em>Full-Duplex Interaction: AI Gains Autonomous Decision-Making</em></strong></p>
<div class="lb-trans"><p><strong><em>1.</em></strong> <strong><em>全双工互动：人工智能获得自主决策能力</em></strong></p>
</div><p>The new model enables stream prediction for responses, listening, and interruptions. AI autonomously decides when to speak, achieving true end-to-end full-duplex interaction, where AI and users can talk simultaneously (e.g., debating, arguing, singing), appropriately interrupt each other, or proactively break silence to initiate topics.</p>
<div class="lb-trans"><p>新模型实现了响应的流预测、倾听和打断。人工智能自主决定何时发言，实现真正的端到端全双工互动，人工智能和用户可以同时交谈（例如，辩论、争论、唱歌），适当地打断彼此，或主动打破沉默以引入话题。</p>
</div><p>This autonomy allows AI to master interaction timing, significantly enhancing dialogue naturalness and enabling immersive, real-world-like exchanges during extended, multi-turn conversations.</p>
<div class="lb-trans"><p>这种自主性使人工智能能够掌握互动时机，显著增强对话的自然性，并在延续的多轮对话中实现沉浸式、现实生活般的交流。</p>
</div><p><strong><em>2.</em></strong> <strong><em>Colloquial &amp; Emotional Expression: More Vivid Interactions</em></strong></p>
<div class="lb-trans"><p><strong><em>2.</em></strong> <strong><em>口语化与情感表达：更生动的互动</em></strong></p>
</div><p>To make AI interactions feel more human-like, the model achieves comprehensive enhancements across multiple dimensions, including emotional expression, vocal characteristics, and conversational content, bringing it closer to real-world communication.</p>
<div class="lb-trans"><p>为了让人工智能的互动更具人性化，该模型在多个维度上实现了全面增强，包括情感表达、声音特征和对话内容，使其更接近现实世界的沟通。</p>
</div><p>For instance, in emotional expression, beyond foundational capabilities like laughter, crying, or anger, the upgraded model delivers more pronounced vocal fluctuations that evolve naturally with the conversation. Its pronunciation now incorporates organic speech elements such as filler words, occasional stammering, common catchphrases, coughs, and other everyday vocal nuances. Furthermore, AI-generated dialogue leans into colloquial and socially fluid language rather than rigid, written-language patterns.</p>
<div class="lb-trans"><p>例如，在情感表达方面，除了基础能力如笑声、哭泣或愤怒外，升级后的模型提供了更明显的声音波动，随着对话自然演变。其发音现在融入了有机的语音元素，如填充词、偶尔的口吃、常见的口头禅、咳嗽和其他日常语音细微差别。此外，人工智能生成的对话倾向于使用口语化和社会流动的语言，而不是僵硬的书面语言模式。</p>
</div><p><strong><em>3.</em></strong> <strong><em>Contextual Awareness: Time, Events, Environment</em></strong></p>
<div class="lb-trans"><p><strong><em>3.</em></strong> <strong><em>情境意识：时间、事件、环境</em></strong></p>
</div><p>Built on a pure autoregressive architecture with unified text/audio generation (Unified Model), the model leverages strong LLM capabilities to integrate persona, time, environment, and contextual dialogue into AI responses. This allows perceptive, understanding AI to better shape &#34;digital personalities&#34;, create rich storylines, and transform interactions into genuine &#34;exchanges of emotion and information&#34;.</p>
<div class="lb-trans"><p>基于纯自回归架构与统一文本/音频生成（统一模型），该模型利用强大的 LLM 能力将个性、时间、环境和情境对话整合到人工智能的响应中。这使得具有感知能力和理解能力的人工智能能够更好地塑造 “数字个性”，创造丰富的故事情节，并将互动转变为真正的 “情感和信息的交流”。</p>
</div><p>Notably, Soul's AI team is currently exploring how to extend its full-duplex voice call model to multi-person scenarios. For example, in group voice conversations, the AI leverages its autonomous decision-making capability to identify optimal speaking moments, effectively facilitate topic discussions and extensions, and seamlessly integrate into authentic social dynamics as an active participant.</p>
<div class="lb-trans"><p>值得注意的是，Soul 的人工智能团队目前正在探索如何将其全双工语音通话模型扩展到多人场景。例如，在群体语音对话中，人工智能利用其自主决策能力识别最佳发言时机，有效促进话题讨论和扩展，并作为积极参与者无缝融入真实的社交动态。</p>
</div><p><strong>Integrating AI into Social Networks: Delivering Emotional and Informational Value</strong></p>
<div class="lb-trans"><p><strong>将人工智能融入社交网络：提供情感和信息价值</strong></p>
</div><p>Drawing on deep insights into social dynamics, Soul rapidly implements technology at the application layer and refines products based on user feedback. Its AI roadmap focuses on two paths: &#34;AI helping users make friends (AI-assisted socializing)&#34; and &#34;AI being friends with users (human-AI interaction)&#34;. Corresponding features, such as the emotionally intelligent &#34;AI Companion&#34; and chat-assistant &#34;AI Chat Assistant&#34;, have garnered enthusiastic user adoption.</p>
<div class="lb-trans"><p>基于对社交动态的深刻洞察，Soul 快速在应用层实施技术，并根据用户反馈优化产品。其人工智能路线图专注于两条路径：“人工智能帮助用户交朋友（人工智能辅助社交）” 和 “人工智能与用户成为朋友（人机互动）”。相应的功能，如情感智能的 “人工智能伴侣” 和聊天助手 “人工智能聊天助手”，已获得用户的热烈采用。</p>
</div><p>Human-AI interaction specifically aims to deliver emotional and informational value through human-like capabilities. According to Soul's Just So Soul Institute (2025 Gen Z AI Usage Report, March 2025; sample: 3,680):</p>
<div class="lb-trans"><p>人机互动特别旨在通过类人能力提供情感和信息价值。根据 Soul 的 Just So Soul Institute（2025 年 Z 世代人工智能使用报告，2025 年 3 月；样本：3680）：</p>
</div><ul>
<li>Nearly 40% of young people use AI daily for emotional companionship.</li>
<li>71.1% are willing to befriend AI (vs. 32.8% in the 2024 Gen Z AIGC Attitudes Report).</li>
</ul>
<div class="lb-trans"><ul>
<li>几乎 40% 的年轻人每天使用人工智能进行情感陪伴。</li>
<li>71.1% 愿意与人工智能交朋友（相比于 2024 年 Z 世代 AIGC 态度报告中的 32.8%）。</li>
</ul>
</div><p>As AI reshapes Generation Z's perception of social relationships, it also gives rise to new demands. A survey conducted by Soul among active users of &#34;AI Companions&#34; reveals that approximately 60% desire more human-like AI behavior.</p>
<div class="lb-trans"><p>随着人工智能重塑 Z 世代对社交关系的认知，它也带来了新的需求。Soul 对 “人工智能伴侣” 的活跃用户进行的调查显示，约 60% 的人希望人工智能表现得更像人类。</p>
</div><p>This full-duplex model upgrade significantly advances AI's interactive abilities, infusing presence and emotional warmth into human-AI exchanges, and will propel AI socializing into a new era.</p>
<div class="lb-trans"><p>这一全双工模型的升级显著提升了人工智能的互动能力，为人机交流注入了存在感和情感温度，将推动人工智能社交进入一个新纪元。</p>
</div><p>As a platform rooted in authentic human connections, Soul is committed to building a social ecosystem where &#34;AI Beings&#34; and &#34;Human Beings&#34; coexist. By continuously investing in cutting-edge technology, Soul empowers AI to enrich users' emotional support systems, diversify emotional experiences, and ultimately enhance individual well-being and belonging.</p>
<div class="lb-trans"><p>作为一个根植于真实人际关系的平台，Soul 致力于构建一个&#34;AI 存在&#34;与&#34;人类&#34;共存的社交生态系统。通过持续投资于前沿技术，Soul 使人工智能能够丰富用户的情感支持系统，丰富情感体验，最终提升个人的幸福感和归属感。</p>
</div><p>SOURCE Soul App</p>
<div class="lb-trans"><p>来源 Soul App</p>
</div><p><img src="https://imageproxy.pbkrs.com/https://rt.prnewswire.com/rt.gif/query-RGF0ZUlkPTIwMjUwNzEwJk5ld3NJdGVtSWQ9Q04yNzc2MSZUcmFuc21pc3Npb25fSWQ9MjAyNTA3MTAwMTIzUFJfTkVXU19VU1BSX19fX19DTjI3NzYx" alt="" original-src="https://rt.prnewswire.com/rt.gif?NewsItemId=CN27761&amp;Transmission_Id=202507100123PR_NEWS_USPR_____CN27761&amp;DateId=20250710"/></p>

Soul App 推出革命性的全双工语音模型：AI 实现自主语音时序控制