--- title: LangSmith now traces voice agents from Pipecat, LiveKit,... description: LangSmith now supports tracing for voice agents built with Pipecat, LiveKit, OpenAI Realtime, and Gemini Live (via Google ADK). Each trace captures full... canonical: https://daily.dev/posts/langsmith-now-traces-voice-agents-from-pipecat-livekit-openai-realtime-and-gemini-live-fsywjie8d twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: LangSmith now traces voice agents from Pipecat, LiveKit, OpenAI Realtime, and Gemini Live | daily.dev og:description: LangSmith now supports tracing for voice agents built with Pipecat, LiveKit, OpenAI Realtime, and Gemini Live (via Google ADK). Each trace captures full... og:url: https://daily.dev/posts/langsmith-now-traces-voice-agents-from-pipecat-livekit-openai-realtime-and-gemini-live-fsywjie8d og:image: https://api.daily.dev/og/posts/FsywjiE8D.png og:image:alt: LangSmith now traces voice agents from Pipecat, LiveKit, OpenAI Realtime, and Gemini Live og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # LangSmith now traces voice agents from Pipecat, LiveKit, OpenAI Realtime, and Gemini Live **[Collections](https://daily.dev/sources/collections)** · 2 min read · 1 upvotes · 0 comments ## Summary LangSmith now supports tracing for voice agents built with Pipecat, LiveKit, OpenAI Realtime, and Gemini Live (via Google ADK). Each trace captures full conversation audio, STT/TTS latency, interruptions, tool calls, and voice activity detection events in a single trace tree. Both classic STT→LLM→TTS pipelines and direct speech-to-speech architectures are supported. Integration requires only a few lines of Python. ## Content LangSmith added tracing support for voice agents across four major frameworks: Pipecat, LiveKit, OpenAI Realtime, and Gemini Live via the Google ADK. Voice agents have historically been hard to debug. Audio flows through multiple stages - speech detection, transcription, LLM inference, synthesis - and when something goes wrong, there's rarely a clean log to look at. This integration pulls all of that into a single trace tree. ## What gets captured Each trace includes: - Full conversation audio - STT and TTS latency - Interruption events (and exactly which audio the user heard before interrupting) - Tool calls with timing - for example, a weather lookup that took 1.29 seconds - Voice activity detection events - Token usage, with native audio tokens separated from regular text tokens The timeline view color-codes user speech in blue, model responses in orange, and custom tool executions separately, so it's easy to see what happened and when. Both "sandwich" architectures (STT → LLM → TTS) and native speech-to-speech setups like Gemini Live are supported, so teams get the same debugging workflow regardless of how their pipeline is built. ## Setup Each integration takes a few lines of Python. For Gemini Live via the Google ADK, register `LangSmithGoogleADKLivePlugin` in your agent and you're done. ## Similar posts on daily.dev - [Building Your First Voice Agent The Sound of the Future: How to Build an AI Voice Agent](https://daily.dev/posts/building-your-first-voice-agent-the-sound-of-the-future-how-to-build-an-ai-voice-agent-dmfmzti5q) · The Miners · 2 upvotes · 0 comments --- Tags: [#python](https://daily.dev/tags/python), [#ai-agents](https://daily.dev/tags/ai-agents), [#observability](https://daily.dev/tags/observability), [#langsmith](https://daily.dev/tags/langsmith) [View this post on daily.dev](https://daily.dev/posts/langsmith-now-traces-voice-agents-from-pipecat-livekit-openai-realtime-and-gemini-live-fsywjie8d) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"LangSmith now traces voice agents from Pipecat, LiveKit, OpenAI Realtime, and Gemini Live","url":"https://daily.dev/posts/langsmith-now-traces-voice-agents-from-pipecat-livekit-openai-realtime-and-gemini-live-fsywjie8d","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/langsmith-now-traces-voice-agents-from-pipecat-livekit-openai-realtime-and-gemini-live-fsywjie8d"},"datePublished":"2026-07-21T17:30:42.602Z","dateModified":"2026-07-23T14:34:12.367Z","description":"LangSmith now supports tracing for voice agents built with Pipecat, LiveKit, OpenAI Realtime, and Gemini Live (via Google ADK). Each trace captures full...","isAccessibleForFree":true,"articleSection":"Collections","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"Collections","logo":"https://media.daily.dev/image/upload/s--fk_6ycEi--/f_auto,q_auto/v1780996001/logos/collections?_a=BAMAMiWQ0","url":"https://daily.dev/sources/collections"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/langsmith-now-traces-voice-agents-from-pipecat-livekit-openai-realtime-and-gemini-live-fsywjie8d","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":1},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"python,ai-agents,observability,langsmith","timeRequired":"PT2M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Collections","item":"https://daily.dev/sources/collections"},{"@type":"ListItem","position":3,"name":"LangSmith now traces voice agents from Pipecat, LiveKit, OpenAI Realtime, and Gemini Live"}]} ```