jpatel98 / transcription-server

Server-side repo for a transcription app that converts user inputted audio to text using a machine learning API.

Transcription App - Server

The app uses Assembly AI's API to convert user inputted audio to text, making it easier for everyone from journalists to students to transcribe speech and make notes.

NOTE: This app needs the transcription-client to run in parallel for full functionality

Backend Stack

Run Locally

Clone the project

  git clone [email protected]:jpatel98/transcription-server.git

Go to the project directory

  cd transcription-server

Install dependencies

  npm install

Start the app

  npm run start

Run in a dev environment

  npm run dev

Environment Variables

To locally run this project, you will need to add the following env variables and configure your AWS S3 bucket. Here is a good walkthrough to configure your S3 bucket.

AWS_BUCKET_NAME

AWS_BUCKET_REGION

AWS_ARN

AWS_ACCESS_KEY

AWS_SECRET_KEY

ASSEMBLY_API_KEY

Lessons Learned

Lessons learned while building this project.

• Divide the problem in smaller chunks.
• Learn to get better at time management.
• Read documentation before diving in.
• Simple is better.

Roadmap

• Connect to a database and allow users to save their transcripts.
• Enable video transcription and implement a video player on screen.
• Create a transcript editor using Draft.js.
• Add realtime transcription feature.