BugSpeak - Speak a bug, ship issue

BugSpeak helps developers and QA file cleaner GitHub issues faster by speaking instead of typing. Using the AssemblyAI Voice Agent API, the agent listens in real time, asks only the minimum clarifying questions, then calls a client-side tool to create a structured GitHub issue. Each issue includes a clear title, summary, steps to reproduce, expected vs actual behavior, environment details, severity, and labels. The app is built with Next.js. Audio is streamed over a WebSocket to AssemblyAI, which handles speech-to-text, reasoning, and spoken replies in one Voice Agent session. Tool calling connects the conversation to GitHub so the demo ends with a real filed issue. This solves a common workflow pain: bug reports often arrive messy from chat or voice notes. BugSpeak turns that into an actionable issue ready for engineers. It showcases AssemblyAI Voice Agent API strengths—low-latency conversation, turn-taking, and tool use—in a practical product for software teams.