<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Voice Agent on Text Matrix</title><link>https://txtmix.com/tags/voice-agent/</link><description>Recent content in Voice Agent on Text Matrix</description><generator>Hugo</generator><language>zh-cn</language><lastBuildDate>Tue, 21 Jul 2026 20:06:14 +0800</lastBuildDate><atom:link href="https://txtmix.com/tags/voice-agent/index.xml" rel="self" type="application/rss+xml"/><item><title>speech-to-speech：Hugging Face 开源的 OpenAI Realtime 兼容语音 Agent 全栈 pipeline</title><link>https://txtmix.com/posts/tech/huggingface-speech-to-speech-voice-agent-pipeline-guide/</link><pubDate>Tue, 07 Jul 2026 03:00:07 +0800</pubDate><guid>https://txtmix.com/posts/tech/huggingface-speech-to-speech-voice-agent-pipeline-guide/</guid><description>&lt;h1 id="speech-to-speech一个能本地化的-openai-realtime-兼容语音-agent">speech-to-speech：一个能本地化的 OpenAI Realtime 兼容语音 Agent&lt;/h1>
&lt;p>OpenAI 的 Realtime API 把&amp;quot;语音对话&amp;quot;做成了单端 WebSocket 调用——但它强绑 OpenAI 的 STT/TTS/LLM，定价、隐私、定制都受限于供应商。Hugging Face 的 &lt;code>speech-to-speech&lt;/code>（5.4k stars / Apache 2.0）做了一个&lt;strong>接口形态完全兼容 OpenAI Realtime&lt;/strong>、但&lt;strong>每一段模块都能换&lt;/strong>的开源替代。本文拆它的四段模块化设计与本地部署路径。&lt;/p></description></item></channel></rss>