skills.rest
darantrute
darantruteCOMMUNITY
Explore all skills by this author →
GitHub

deepgram

Transcribe pre-recorded and real-time audio using Deepgram's Nova-3 model.

Updated Aug 23, 2026
Search Tags:#next.js#transcription#api integration#websockets#speech-to-text#real-time audio#deepgram
One-click install
npx skills add https://github.com/darantrute/userdecisiontree --skill deepgram
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: deepgram
Source: https://github.com/darantrute/userdecisiontree/tree/main/.claude/skills/deepgram
Command: npx skills add https://github.com/darantrute/userdecisiontree --skill deepgram

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires @deepgram/sdk, and includes references (resource) components.

What problem does it solve?

This Skill simplifies the integration of Deepgram's advanced speech-to-text capabilities into your applications, enabling accurate transcription of both pre-recorded and real-time audio. It helps overcome the complexities of WebSocket streaming, model selection, and robust error handling, allowing you to focus on building innovative voice-enabled features.

Core Features & Use Cases

  • Real-Time & Pre-recorded Transcription: Seamlessly transcribe live audio streams via WebSockets or process audio files and URLs with high accuracy.
  • Advanced AI Models: Leverages the latest Nova-3 model for multilingual support, smart formatting (currency, dates), and speaker diarization.
  • Secure Next.js Integration: Provides patterns for securely handling API keys on the server-side and implementing client-side audio capture with a WebSocket proxy.
  • Use Case: Build a real-time captioning service for live events, develop a voice-controlled interface for your application, or automatically transcribe meeting recordings for searchable notes.

Quick Start

Use the deepgram skill to transcribe an audio file uploaded via a Next.js API route, ensuring smart formatting and speaker diarization.