Real-Time Speech Transcription Frontend

A Next.js-based speech transcription frontend that integrates with LiveKit for real-time audio transcription. The application requires a companion transcription agent to process audio streams and display transcribed text. Built with TypeScript and React, it demonstrates a minimal frontend implementation for LiveKit's transcription capabilities, typically paired with backend agents using Groq for speech-to-text processing.

No demo yetView source Fork template

The numbers

latency—

cost / min—

frameworkLiveKit Agents

The stack

telephonyWeb Only

speech-to-textDeepgram Nova-3

llmGroq Llama 3.3 70B

text-to-speechOpenAI TTS-1

System prompt

No prompt published.

Config

config.json

{
  "sdk": [
    "livekit-client",
    "livekit-server-sdk"
  ],
  "frontend": [
    "nextjs",
    "react",
    "typescript",
    "tailwind"
  ],
  "package_manager": "pnpm"
}

Voice AI recipes, picks, and analysis.

Get the useful new templates plus the occasional teardown of what’s working in production voice AI.