From 6641bf28527921fd2a3e9747edf586ceb11e1ff6 Mon Sep 17 00:00:00 2001 From: Bearclaw Date: Wed, 18 Mar 2026 13:20:52 +0000 Subject: [PATCH] Add voice agent routing callout to streaming quickstart Evaluators building voice agents follow the streaming quickstart, use generic defaults, and get suboptimal results. This adds a prominent tip after the Overview section directing them to the Voice Agent Best Practices guide for optimized model selection, prompting, and turn detection configuration. Co-Authored-By: Claude Opus 4.6 --- fern/pages/getting-started/transcribe-streaming-audio.mdx | 4 ++++ 1 file changed, 4 insertions(+) diff --git a/fern/pages/getting-started/transcribe-streaming-audio.mdx b/fern/pages/getting-started/transcribe-streaming-audio.mdx index 52bb699d..f74c7854 100644 --- a/fern/pages/getting-started/transcribe-streaming-audio.mdx +++ b/fern/pages/getting-started/transcribe-streaming-audio.mdx @@ -16,6 +16,10 @@ description: Quickstart guide for transcribing real-time streaming audio from a By the end of this tutorial, you'll be able to transcribe audio from your microphone. + +**Building a voice agent?** This quickstart covers general-purpose streaming transcription. For voice agents, we strongly recommend using Universal-3 Pro Streaming with optimized turn detection and prompting. See our [Voice Agent Best Practices](/docs/voice-agent-best-practices) guide for model selection, configuration, and integration patterns that deliver the best conversational AI experience. + + ## Before you begin To complete this tutorial, you need: