Skip to content

tanmaynaik/transvox

Repository files navigation

TransVox

Record your speech, convert it to text, and display it in real-time.


Overview

TransVox is an experimental project that allows users to record speech from the browser and send it to a backend for transcription.
The backend uses AssemblyAI to convert audio into text, which is then displayed in the UI textbox.


Features

  • Record Speech from the browser using MediaRecorder
  • Speech-to-Text (STT) via AssemblyAI
  • 🚧 Text Translation (powered by Murf AI)
  • 🚧 Text-to-Speech Playback using Murf AI voices
  • Modern UI built with React + Tailwind CSS

Tech Stack

Layer Technology
Frontend React, Tailwind CSS, Vite
Backend Flask (Python)
Speech-to-Text AssemblyAI
Translation Murf AI Translation API
Text-to-Speech Murf AI TTS API
Audio Handling pydub + FFmpeg

Getting Started

Prerequisites

  • Node.js & npm (for frontend)
  • Python 3.8+ (for backend)
  • FFmpeg installed (for audio processing)
  • API keys:
    • AssemblyAI (for STT)
    • Murf AI (for Translation + TTS)

Frontend Setup

cd frontend
npm install
npm run dev

Backend Setup

cd backend
python -m venv venv
source venv/bin/activate      # macOS/Linux
venv\Scripts\activate         # Windows

python app.py

Acknowledgements

-Built as part of the Murf AI Hackathon/Contest 🏆

-Thanks to Murf AI for their Translation & Text-to-Speech APIs

-Uses AssemblyAI for transcription

About

Built as part of the Murf AI Coding Challenge

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors