deepgram
Virgin bootstrap template for Claude Code system with Next.js 15, React 19, TypeScript 5, Prisma, Clerk, Deepgram, Groq, and Llama 3.3 70B
- Category: discovered
- Tags: github, community
- Version: 1.0.0
- Author: walid
- Downloads: 0
- Updated: 2025-11-22
Download the latest version (v1.0.0) from this page.
SKILL.md
# Deepgram SDK Skill This skill provides patterns and best practices for using Deepgram's speech-to-text API with Next.js applications. ## What This Skill Covers - **Pre-recorded Audio Transcription** - Transcribe audio files and URLs - **Real-Time Streaming** - WebSocket-based live transcription - **Nova-3 Model** - Latest model with multilingual support and <300ms latency - **Smart Formatting** - Auto-format currency, dates, phone numbers - **Speaker Diarization** - Separate and identify different speakers - **Error Handling** - Common mistakes and how to avoid them - **Next.js Integration** - API routes, WebSocket proxies, React hooks ## When to Use This Skill Use this skill when working on: - Voice-to-text features - Meeting transcription - Real-time captioning - Voice command interfaces - Podcast/video transcription - Multi-speaker audio analysis ## Key Features - **Server-Side Only** - Properly secure API keys - **WebSocket Patterns** - Real-time streaming with KeepAlive - **Error Recovery** - Handle connection issues and buffering - **Latest Models** - Nova-3 with 90%+ accuracy - **TypeScript Support** - Fully typed SDK integration ## Related Skills - **nextjs** - For API route patterns - **react** - For client-side audio capture - **typescript** - For type-safe implementation ## Documentation See `references/external/deepgram-patterns.md` for comprehensive patterns, examples, and troubleshooting. --- **Skill Version:** 1.0.0 **Last Verified:** November 12, 2025 **Next Review:** December 2025