whispey

Heads-down with the team building Whispey SDK for LiveKit voice agents — sharing a small win that actually saves time in production. We turned LiveKit telemetry into something you can act on: → Session overview (turns, traces, bugs) → Waterfall timeline across STT → LLM → TTS (with each OpenTelemetry span broken down in a structured format) → Click any span → side sheet with full request + raw JSON (super helpful for debugging at the atomic level) → P75 and P50 latencies surfaced at a glance The setup “aha”? Just one flag: enable_otel=True 94 spans → Structured insights you can actually ship with. LiveKit gives the data. Whispey makes it production-ready. If you’re running LiveKit agents in prod — what’s the one insight that would save you the most debugging time right now?

I have built this open source tool to make the bug reporting seamless in voice AI agents. This has been very helpful for me in perfecting the Voice Agents I build for our customers. Initially, I used to give the voice agent to our customers for their testing and feedback. And they used to comeback to us after a few days with all their feedback written on multiple documents/papers. Rarely they used to mention time stamp etc.,. making it very difficult for me to pin point the issue. Now which this tool integrated in my Voice Agents, they can simply say "Feedback Start" --> Give feedback --> Say "Feedback Stop". And continue talking to the agent. In the backend I can clearly see where user wants improvement and at which conversation turn they felt that way. I want to share this tool with a wider community. Pls feel free to use it and contrib

Github - github.com/pype-ai-main/whispey See it working - https://youtu.be/BMAu-sRLp2M

We run multiple voice AI agents on LiveKit and kept running into the same headaches: * No way to see TTFT (time to first token) per call * Hard to measure cost per call * No visibility into latency across STT → LLM → TTS * Debugging was basically guesswork So we built Whispey to solve it. What it does: * Cost analytics * Latency breakdowns (VAD → STT → LLM → TTS) * Agent evaluation metrics * Campaign tracking * Downloadable recordings + transcripts Runs on Vercel + Supabase today, AWS-native version is coming. It’s tightly integrated with LiveKit. https://www.whispey.xyz/ https://github.com/PYPE-AI-MAIN/whispey If you’ve built voice agents on LiveKit, we’d love for you to try it out and share what’s been working for you. Always looking for best practices from other builders.