Skip to content

Real-time captions with optional translation using Next.js.

Notifications You must be signed in to change notification settings

rndstan/captions

This branch is up to date with paratustra/captions:main.

Folders and files

NameName
Last commit message
Last commit date

Latest commit

cd2230a · Jul 1, 2024

History

1 Commit
Jul 1, 2024
Jul 1, 2024
Jul 1, 2024
Jul 1, 2024
Jul 1, 2024
Jul 1, 2024
Jul 1, 2024
Jul 1, 2024
Jul 1, 2024
Jul 1, 2024
Jul 1, 2024
Jul 1, 2024
Jul 1, 2024
Jul 1, 2024

Repository files navigation

Real-time captions

This project implements a real-time speech-to-text transcription service with optional translation capabilities using React, Next.js, Deepgram for speech recognition, and Groq for translation.

Features

  • Real-time speech-to-text transcription
  • Optional automatic translation to English
  • Toggle switch for enabling/disabling translation

Technologies Used

  • React
  • Next.js
  • Deepgram API for speech recognition
  • Groq API for translation
  • Framer Motion for animations
  • Tailwind CSS for styling

Setup

  1. Clone the repository
  2. Install dependencies: pnpm install
  3. Set up environment variables:
    • DEEPGRAM_API_KEY: Your Deepgram API key
    • GROQ_API_KEY: Your Groq API key
  4. Run the development server: pnpm run dev

API Routes

  • /api/authenticate: Provides the Deepgram API key to the client
  • /api/translate: Handles text translation using the Groq API

Usage

  1. Start the application
  2. Grant microphone access when prompted
  3. Speak into the microphone to see real-time transcription
  4. Toggle the switch to enable/disable automatic translation to English

Notes

  • The transcription is set up for Spanish (es-ES) by default. Modify the language parameter in connectToDeepgram to change the input language.
  • The translation service uses Groq's gemma-7b-it model for English translation.

About

Real-time captions with optional translation using Next.js.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • JavaScript 91.2%
  • CSS 8.8%