Transcript Seeker

Open-source Transcription Playground with an MIT License

Transcript Seeker is an open-source transcription playground powered by transcription APIs and the meeting bot API of Meeting Baas 🐟.

With Transcript Seeker, you can easily upload audio or video files, record a video-meeting and transcribe your content. The interface then allows you to:

Visualize transcripts synced with your media
Navigate through content by clicking on specific words
Use speaker diarization from transcription APIs or the Meeting Baas 🐟 bot API
Access additional features like summarization and named entity recognition

As Transcript Seeker is open-source, you can customize and extend those functionalities. Feel free to check out the to-do list on the GitHub repository, to contribute or see what's coming next.

Features

Synchronized Playback: Click on any word in the transcript to jump to that point in the recording.
AI Interaction: Chat with transcripts using an OpenAI integration and Large Language Models (LLMs).
Annotation: Add notes to recordings manually or automatically with AI summaries provided by the APIs you're using.
File Transcription: Upload video/audio recordings and transcribe using the transcription API of your choice. Compatible APIs include: Whisper, Gladia, Deepgram and AssemblyAI.
Meeting Recording: Generate recordings with metadata using the Meeting Bot as a Service (Baas) 🐟 API for video-meetings on: Zoom, Google Meet, and Microsoft Teams.
File Management: Upload, load and store local or distant file recordings.

Deployment Options:

Browser-based setup:
- Runs directly in your browser
- Uses IndexedDB for local storage
- Provides an easy and quick setup
- Offers offline functionality in standalone mode

Server-based setup:
- Option to run with an Express-JS backend
- Easy to deploy and scale

Tech Stack:

Frontend: React, Vite, TypeScript, TailwindCSS, IndexedDB
Media Playback: Vidstack
Optional Backend: Express-JS

To launch Transcript Seeker, first make sure you have Git, Node.js, and Yarn installed.
Then, copy the script below, save it as run.sh and make it executable:


 
chmod +x run.sh
./run.sh

View on GitHub
 
#!/bin/bash
# 🚀 Quick Start for Transcript Seeker 🚀
# Save this file as run.sh, chmod +x run.sh, and run it
# ⚠️ Make sure you have Git, Node.js, and Yarn installed before proceeding!

# clone and navigate to the repo
git clone https://github.com/Meeting-Baas/transcript-seeker && cd transcript-seeker

# Install dependencies
echo "🔧 Installing dependencies..."
yarn install && yarn install:all || { echo "❌ Error: Failed to install dependencies"; exit 1; }

echo "🏗️ Building dependencies..."
yarn run build || { echo "❌ Error: Failed to build dependencies"; exit 2; }

echo "📄 Copying .env file..."
cp .env.example .env || { echo "⚠️ Warning: Failed to copy .env file. You may need to set up environment variables manually."; }

echo "🚀 Starting the development server..."
cd client && yarn run dev || { echo "❌ Error: Failed to start the development server"; exit 3; }