Can I use this for live video calls?

Absolutely. It is specifically optimized for live streaming transcription with a dedicated React hook and real-time UI components for scrolling captions.

What are the technical prerequisites?

You need a Next.js App Router project with the video-room (LiveKit), db (Drizzle), and env-config skills already installed.

Does this support multiple speakers in a single room?

Yes, it includes built-in speaker diarization support to automatically identify, label, and distinguish between different participants in the transcript.

Where are the transcripts stored after a meeting?

Transcripts and individual segments are persisted to your PostgreSQL database using Drizzle ORM, allowing for easy search and historical retrieval.

Which AI service does this skill use for transcription?

This skill utilizes Deepgram's Nova-3 model, providing high-accuracy, low-latency speech-to-text for both live and batch processing.

AI Video Transcription

Name: AI Video Transcription
Author: mattwoodco

bymattwoodco

•

Herramientas de Colaboración

Implements real-time and batch speech-to-text with Deepgram for video applications, featuring live captions and persistent transcript storage.

This skill integrates Deepgram's advanced speech-to-text capabilities into Next.js applications, specifically designed to work with LiveKit video rooms. It provides a comprehensive solution for both live streaming transcription with real-time overlays and batch processing of recorded audio. The skill includes automatic speaker diarization to label different participants, a searchable post-meeting transcript viewer, and robust database persistence using Drizzle ORM and PostgreSQL to ensure all meeting data is archived and accessible.

Características Principales

01Live transcript UI components with scrolling overlays and speaker labels

02Batch processing for pre-recorded audio with speaker diarization

033 GitHub stars

04Real-time streaming transcription via Deepgram WebSocket API

05Database persistence for transcripts using Drizzle ORM and Postgres

06Full post-meeting transcript viewer with search and filtering

Casos de Uso

01Building automated speech-to-text workflows for uploaded audio assets

02Adding live accessibility captions to video conferencing applications

03Generating searchable meeting archives and minutes for internal tools

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add mattwoodco/skills transcription

For use in Claude.ai and ChatGPT

Download Skill